2023-09-27_HD-SNN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ad4f5829-d4e8-499e-b815-50eda50f01c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "datetag = None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "0eac0b09",
   "metadata": {},
   "outputs": [],
   "source": [
    "datetag = '2023-09-27'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "27340972",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "datetag='2023-09-27'\n"
     ]
    }
   ],
   "source": [
    "if datetag is None:\n",
    "    from time import strftime, gmtime\n",
    "    datetag = strftime(\"%Y-%m-%d\", gmtime())\n",
    "print(f\"datetag='{datetag}'\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b959c09c-3907-46cd-8c06-303c59af8597",
   "metadata": {
    "tags": []
   },
   "source": [
    "# HD-SNN: Accurate detection of spiking motifs in multi-unit raster plots"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf094a22",
   "metadata": {},
   "source": [
    "\n",
    "We design a model based on heterogeneous delays. We will define in the assembly of neurons where each synapse\n",
    "is defined by a weight *ws* at different delays *τs*. See the following figure: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "65bf0ef2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# %pip install -U pip pandas\n",
    "# uncomment for MacOS with silicon M1\n",
    "# %pip install -U --pre torch torchvision --extra-index-url https://download.pytorch.org/whl/nightly/cpu"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "6486aa33",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "device(type='mps')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import torch\n",
    "torch.set_printoptions(precision=3, linewidth=140, sci_mode=False)\n",
    "\n",
    "if torch.backends.mps.is_available():\n",
    "    device = torch.device('mps')\n",
    "elif torch.cuda.is_available():\n",
    "    device = torch.device('cuda')\n",
    "else:\n",
    "    device = torch.device('cpu')\n",
    "# device = torch.device('cpu') # uncomment to force CPU usage\n",
    "device"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c940312f",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "phi = np.sqrt(5)/2 + 1/2\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import matplotlib\n",
    "subplotpars = matplotlib.figure.SubplotParams(left=0.125, right=.95, bottom=0.25, top=.975, wspace=0.05, hspace=0.05,)\n",
    "\n",
    "figpath = 'figures'\n",
    "figpath = None\n",
    "\n",
    "def printfig(fig, name, exts=['pdf', 'svg'], figpath=figpath, dpi_exp=None, bbox='tight'):\n",
    "    for ext in exts:\n",
    "        fig.savefig(os.path.join(figpath, name + '.' + ext), dpi=dpi_exp, bbox_inches=bbox, transparent=True)\n",
    "\n",
    "if not(figpath is None): os.makedirs(figpath, exist_ok=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d4147459-e8c8-40b8-a44b-862186f0533d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# https://docs.python.org/3/library/dataclasses.html?highlight=dataclass#module-dataclasses\n",
    "from dataclasses import dataclass, asdict, field\n",
    "\n",
    "@dataclass\n",
    "class Params:\n",
    "    datetag: str = datetag\n",
    "    N_pre: int = 128 # number of presynaptic inputs\n",
    "    N_SM_time: int = 31 # number of timesteps in SM, must be a odd number for convolutions\n",
    "    N_SMs: int = 233 # number of spiking motifs\n",
    "    E_SM: float = 4. # excitability range of SMs\n",
    "    p_SM: float = .05 # ratio of non-zero coefficients in SMs\n",
    "    T_trial: float = 1. # time duration of a trial (in second)\n",
    "    tau_decay: float = .4 # time constant for the decay of the SM enveloppe (in second)\n",
    "    tau_rise: float = .05 # time constant for the rise of the SM enveloppe (in second)\n",
    "\n",
    "    ## Raster plots\n",
    "    N_time: int =  2**10 # number of timebins for the raster plot\n",
    "    N_trials: int = 10 # number of trials\n",
    "    p_B: float = .001 # prior probability of firing for postsynaptic raster plot (spike per timebin)\n",
    "    p_A: float = .001 # prior probability of firing for presynaptic raster plot (spike per timebin)\n",
    "    seed: int = 42 # seed\n",
    "\n",
    "    ## figures\n",
    "    verbose: bool = False # Displays more verbose output.\n",
    "    fig_width: float = 12 # width of figure\n",
    "    phi: float = 1.61803 # beauty is gold\n",
    "    N_SM_show: int = 5 # number of SM to show in plot_SM\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "b8daaf2b-0466-411b-a6ba-6b5f785d102a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Params(datetag='2023-09-27', N_pre=128, N_SM_time=31, N_SMs=233, E_SM=4.0, p_SM=0.05, T_trial=1.0, tau_decay=0.4, tau_rise=0.05, N_time=1024, N_trials=10, p_B=0.001, p_A=0.001, seed=42, verbose=False, fig_width=12, phi=1.61803, N_SM_show=5)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "opt = Params()\n",
    "opt"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "9047cb89-b0b1-42ba-b4c0-d443983a52f2",
   "metadata": {
    "tags": []
   },
   "source": [
    "### generative & inference models for spiking motifs (SM)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "72cfb669-0306-4ec4-b86a-05d3a642e5e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "class ABCD:\n",
    "\n",
    "    def __init__(self, opt):\n",
    "        self.opt: Params = opt\n",
    "        self.init()\n",
    "\n",
    "    def init(self):\n",
    "        \"\"\"\n",
    "        \n",
    "        initializes \"temporal mod\", which governs the overall structure of the motif. \n",
    "        The shape is the product  of two exponenential functions creating a \"psp-like\" \n",
    "        shape. in set_SM the temporal mod will cause each SM to have the \n",
    "        distribution shape of temporal mod, such that the probability distribution \n",
    "        of spike occurences has a fast rise and a slow decay.\n",
    "\n",
    "        \"\"\"\n",
    "        self.logit_p_A = torch.logit(torch.tensor(self.opt.p_A))\n",
    "\n",
    "        temporal_mod = torch.zeros(self.opt.N_SM_time)\n",
    "        self.time = torch.linspace(0, self.opt.T_trial, self.opt.N_time)\n",
    "        self.SM_time = torch.linspace(0, self.opt.T_trial, self.opt.N_SM_time)\n",
    "        temporal_mod = torch.exp(- self.SM_time / self.opt.tau_decay)\n",
    "        temporal_mod *= 1 - torch.exp(- self.SM_time / self.opt.tau_rise)\n",
    "        # temporal_mod = temporal_mod.flip([0])\n",
    "        #self.temporal_mod = (torch.eye(self.opt.N_SMs).unsqueeze(2)) * (temporal_mod.unsqueeze(0).unsqueeze(0))\n",
    "        self.temporal_mod = torch.ones((self.opt.N_pre, self.opt.N_SMs, 1)) * (temporal_mod.unsqueeze(0).unsqueeze(0))\n",
    "        \n",
    "        \"\"\"\n",
    "        torch.eye -> identity matrix \n",
    "        torch.unsqueeze -> wrap in more brackets\n",
    "        torch.flip -> reverses the orders of the elements in the vector\n",
    "        \n",
    "        this creates a 144 by 144 by 3 matrix, \n",
    "        Then at each diagonal there is an identical spike waveform\n",
    "        such that convolving the SM with that matrix will result in a\n",
    "        spike waveform at each presynptic input.\n",
    "        \n",
    "        \"\"\"\n",
    "        spike = torch.tensor([1, -.8, -.2]) # spike shape with depolariation (1) and hyperpolarization ([-.8, -.2])\n",
    "        self.spike = (torch.eye(self.opt.N_SMs).unsqueeze(2)) * (spike.flip([0]).unsqueeze(0).unsqueeze(0))\n",
    "\n",
    "\n",
    "    def set_SM(self, seed=None, seed_offset=0):\n",
    "        \"\"\"\n",
    "        Returns a N_SMs by N_pre by N_SM_time dimensional matrix representing the probability distribution of each elemenet\n",
    "        in the matrix to evoke a spike. Presumably later spikes will be drawn from this distribution via Bernoulli trials.\n",
    "        \"\"\"\n",
    "        if seed is None: seed = self.opt.seed + seed_offset\n",
    "        torch.manual_seed(seed)\n",
    "        \n",
    "        # 1/ define SMs as matrices to be used as kernels\n",
    "        SM = self.opt.E_SM * torch.randn(self.opt.N_pre, self.opt.N_SMs, self.opt.N_SM_time)\n",
    "        #threshold = torch.abs(SM).quantile(1-self.opt.p_SM)\n",
    "\n",
    "        # 2/ zero out everything below the threshold\n",
    "        # TODO : get analytically\n",
    "        from scipy.stats import norm\n",
    "        threshold = self.opt.E_SM * norm.ppf(1-self.opt.p_SM)\n",
    "        SM *= (SM > threshold)\n",
    "\n",
    "        # 3/ modulate in time\n",
    "        SM *= self.temporal_mod\n",
    "\n",
    "        # 4/ convolve with a spike shape to induce some sort of refractory period\n",
    "        SM = torch.conv1d(SM, self.spike, padding=self.spike.shape[-1]//2)\n",
    "        return torch.swapaxes(SM, 0, 1)\n",
    "\n",
    "    def get_b(self, seed=None, seed_offset=1):\n",
    "        \"\"\"\n",
    "        generates a matrix of events for each SM, for each trial, returns a matrix of N_trials, N_SMs, N_time, \n",
    "        draw causes (SMs) as a matrix of sparse SM activations uniformly in postsynaptic space and time\n",
    "        \n",
    "        \"\"\"\n",
    "        if seed is None: seed = self.opt.seed + seed_offset\n",
    "        torch.manual_seed(seed)\n",
    "        \n",
    "        # we set it to zero everywhere\n",
    "        b_proba = torch.zeros(self.opt.N_trials, self.opt.N_SMs, self.opt.N_time)\n",
    "        # set to p_B except outside the borders\n",
    "        b_proba[:, :, (self.opt.N_SM_time//2):-(self.opt.N_SM_time//2)] = self.opt.p_B        \n",
    "        return torch.bernoulli(b_proba)\n",
    "\n",
    "    def plot_raster(self, raster, raster_post=None, SM=None, \n",
    "                    i_trial=0, shift=0, xticks=6, yticks=16, spikelength=.9, \n",
    "                    colors=None, figsize=None, subplotpars=subplotpars, \n",
    "                    ylabel='address', linewidths=1.0):\n",
    "        \"\"\"    \n",
    "        plots b, which is trials by M by T\n",
    "        \"\"\"\n",
    "\n",
    "        N_neurons = raster.shape[1]\n",
    "        if colors is None: # blue if nothing assigned\n",
    "            colors = ['b'] * N_neurons\n",
    "        else: # give the colors or ...\n",
    "            if len(colors)==1: # ... paint everything the same color\n",
    "                colors = colors[0] * N_neurons\n",
    "\n",
    "        if figsize is None: figsize = (self.opt.fig_width, self.opt.fig_width/self.opt.phi)\n",
    "\n",
    "        fig, ax = plt.subplots(1, 1, figsize=figsize, subplotpars=subplotpars)\n",
    "        if raster_post is None:\n",
    "            for i in range(0, N_neurons):\n",
    "                ax.eventplot(np.where(raster[i_trial, i, :] == 1.)[0]+shift, \n",
    "                    colors=colors[i], lineoffsets=1.*i+spikelength/2, \n",
    "                    linelengths=spikelength, linewidths=linewidths)\n",
    "        else:\n",
    "            for i_SM in range(self.opt.N_SMs):\n",
    "                b_ = torch.zeros_like(raster_post)\n",
    "                b_[i_trial, i_SM, :] = raster_post[i_trial, i_SM, :]\n",
    "                a_ = self.draw_a(b_, SM)\n",
    "                for i in range(0, self.opt.N_pre):\n",
    "                    ax.eventplot(np.where(a_[i_trial, i, :] == 1.)[0]+shift, colors=colors[i_SM], lineoffsets=1.*i+spikelength/2, \n",
    "                    linelengths=spikelength, linewidths=linewidths)\n",
    "    \n",
    "        ax.set_ylabel(ylabel)\n",
    "        ax.set_xlabel('Time (a. u.)')\n",
    "        ax.set_xlim(0, self.opt.N_time)\n",
    "        ax.set_ylim(0, N_neurons)\n",
    "\n",
    "        # ax.yaxis.set_minor_locator(matplotlib.ticker.MultipleLocator(yticks))\n",
    "        ax.set_yticks(np.linspace(0, N_neurons, yticks, endpoint=False)+.5)\n",
    "        ax.set_yticklabels(np.linspace(1, N_neurons, yticks, endpoint=True).astype(int))\n",
    "        for side in ['top', 'right']: ax.spines[side].set_visible(False)\n",
    "        \n",
    "        ax.xaxis.set_minor_locator(matplotlib.ticker.MultipleLocator(4))\n",
    "        ax.set_xticks(np.linspace(1, self.opt.N_time, xticks, endpoint=True))\n",
    "        ax.set_xticklabels(np.linspace(1, self.opt.N_time, xticks, endpoint=True).astype(int))\n",
    "\n",
    "        ax.grid(visible=True, axis='y', linestyle='-', lw=.5)\n",
    "        ax.grid(visible=True, axis='x', which='both', linestyle='-', lw=.1)\n",
    "\n",
    "        return fig, ax\n",
    "\n",
    "\n",
    "    def plot_b(self, b, i_trial=0, xticks=6, yticks=16, spikelength=.9, colors=None, figsize=None, subplotpars=subplotpars):\n",
    "        # b_shifted = torch.roll(b, -self.opt.N_SM_time//2, dims=-1)\n",
    "        b_shifted = torch.roll(b, 0, dims=-1) # HACK\n",
    "\n",
    "        fig, ax = self.plot_raster(raster=b_shifted, i_trial=i_trial, xticks=xticks, yticks=yticks, spikelength=spikelength, colors=colors, figsize=figsize, subplotpars=subplotpars, ylabel='@ Motif')\n",
    "        return fig, ax\n",
    "\n",
    "    def model_a_logit(self, b, SM):\n",
    "        \"\"\"\n",
    "\n",
    "        Generative model:\n",
    "        defines the evidence of firing for each presynaptic address over time\n",
    "        returns the logit_p_A rasters for each trial\n",
    "\n",
    "        input b -> postsynaptic raster plot\n",
    "            SM -> spiking motifs as spatio temporal kernels\n",
    "            p_A -> prior proba of firing for the presynaptic addresses\n",
    "        \n",
    "        \"\"\"\n",
    "        logit_p_A_fromB = torch.conv_transpose1d(b*1., SM, padding=self.opt.N_SM_time//2)\n",
    "        return self.logit_p_A + logit_p_A_fromB\n",
    "\n",
    "    def model_a(self, b, SM):\n",
    "        \"\"\"\n",
    "\n",
    "        Generative model:\n",
    "        defines the probability of firing for each presynaptic address over time from its evidence\n",
    "\n",
    "        calls model_a_logit(b,SM) to get the logit, and applies the sigmoid over the logit_p_A rasters to turn the \n",
    "        rasters into probabilities between 1 and 0        \n",
    "        \"\"\"\n",
    "        logit_A = self.model_a_logit(b, SM)\n",
    "        return torch.sigmoid(logit_A)\n",
    "\n",
    "    def draw_a(self, b, SM, seed=None, seed_offset=2):\n",
    "        \"\"\"\n",
    "        calls model_a(b,SM) and generates A_proba which is the probability distribution of neuron spiking events,\n",
    "        then this gets passed through a bernoulli trial, deciding where neurons spike given the probability distribution\n",
    "        \"\"\"\n",
    "        # draws binary events from the probability of firing\n",
    "        if seed is None: seed = self.opt.seed + seed_offset\n",
    "        torch.manual_seed(seed)\n",
    "        # generate the corresponding raster plot\n",
    "        A_proba = self.model_a(b, SM)\n",
    "        return torch.bernoulli(A_proba)\n",
    "\n",
    "    def plot_a(self, a, b=None, SM=None, i_trial=0, shift=0, xticks=6, yticks=16, spikelength=.9, colors=None, figsize=None, subplotpars=subplotpars):\n",
    "        fig, ax = self.plot_raster(raster=a, raster_post=b, SM=SM, i_trial=i_trial, shift=shift,\n",
    "                                   xticks=xticks, yticks=yticks, spikelength=spikelength, colors=colors, \n",
    "                                   figsize=figsize, subplotpars=subplotpars, ylabel='@ Neuron')\n",
    "        return fig, ax\n",
    "\n",
    "    def do_thresholding(self, b, p_B, max_quant=10000000):\n",
    "        \"\"\"\n",
    "\n",
    "        Threshold model:\n",
    "\n",
    "        essentially this function transforms the convoluted matrix which is unnormalized and unbounded into\n",
    "        a matrix of quantiles, and then binarizes the matrix by thresholding. The function returns the \n",
    "        binary b matrix...\n",
    "        \n",
    "        \"\"\"\n",
    "        if len(b.ravel()) > max_quant: # the quantile function does not like having to many samples\n",
    "            ind_quant = torch.randperm(len(b.ravel()))[:max_quant]\n",
    "            b_threshold = torch.quantile(b.ravel()[ind_quant], 1-p_B)\n",
    "        else:\n",
    "            b_threshold = torch.quantile(b, 1-p_B)\n",
    "\n",
    "        b_hat_bin = (b > b_threshold) * 1.\n",
    "        return b_hat_bin\n",
    "\n",
    "    def inference_with_SMs(self, a, b, SM, max_quant=10000000, method='HD-SNN'):\n",
    "        \"\"\"\n",
    "\n",
    "        Detection model:\n",
    "        defines the probability of firing for each postsynaptic address over time\n",
    "        \n",
    "\n",
    "        convolutes matrix a with matrix SM and returns another matrix of the same size as b. Since SM\n",
    "        isn't binary, b_hat also isn't binary... Therefore to turn b_hat back into a binary image the \n",
    "        the function do_thresholding get's called. both b_hat and b_hat_bin get returned.\n",
    "\n",
    "        \"\"\"\n",
    "        # infer \n",
    "        if not(method=='HD-SNN'):\n",
    "            SM = torch.sigmoid(self.logit_p_A +  SM)\n",
    "            # SM = (SM > SM.std())*1. - (SM < -SM.std())*1.\n",
    "\n",
    "        b_hat = torch.conv1d(a*1., SM, padding=self.opt.N_SM_time//2)\n",
    "        # decision\n",
    "        p_B = b.sum()/(self.opt.N_trials*self.opt.N_SMs*self.opt.N_time) \n",
    "        # hack: ça suppose qu'on connait b.sum()...\n",
    "        b_hat_bin = self.do_thresholding(b_hat, p_B, max_quant=max_quant)\n",
    "        return b_hat, b_hat_bin\n",
    "    \n",
    "    def generative_model(self, seed=None, seed_offset=3):\n",
    "        if seed is None: seed = self.opt.seed + seed_offset\n",
    "        torch.manual_seed(seed)\n",
    "        SM, b = self.set_SM(seed=seed), self.get_b(seed=seed+1)\n",
    "        a = self.draw_a(b, SM, seed=seed+2)\n",
    "        return a, b, SM\n",
    "\n",
    "    def test_model(self, SM, SM_true=None, seed=None, seed_offset=4, method='HD-SNN'):\n",
    "        \"\"\"\n",
    "        this function calls inference_with_SMs(a,b,SM). This function calls get_b and draw_a over again, so \n",
    "        if this method is called it will reset the raster plot given SM... ( i think?)... hmmmmm\n",
    "        Or maybe SM are the priors?   \n",
    "        \n",
    "        \"\"\"\n",
    "        if seed is None: seed = self.opt.seed + seed_offset\n",
    "        torch.manual_seed(seed)\n",
    "        if SM_true is None: SM_true = SM\n",
    "        # define SMs\n",
    "        # draw causes (SMs)\n",
    "        b = self.get_b(seed=seed)\n",
    "        # generate the corresponding raster plot\n",
    "        a = self.draw_a(b, SM_true, seed=seed+1)\n",
    "        # infer \n",
    "        _, b_hat_bin = self.inference_with_SMs(a, b, SM, method=method)\n",
    "        # count\n",
    "        accuracy = torch.mean((b_hat_bin == b)*1.)\n",
    "        TP = torch.mean(b_hat_bin[b==1]*1.)\n",
    "        TN = 1-torch.mean(b_hat_bin[b==0]*1.)\n",
    "        return accuracy, TP, TN\n",
    "        \n",
    "    def plot_SM(self, SM, cmap='seismic', colors=None, aspect=None, figsize=None, subplotpars=subplotpars, N_SM_show: int=0):\n",
    "\n",
    "        if N_SM_show == 0: N_SM_show = self.opt.N_SM_show\n",
    "        if SM.dtype == torch.bool:\n",
    "            SM_max = 1\n",
    "            SM_min = 0\n",
    "            cmap = 'binary'\n",
    "        else:\n",
    "            # SM = SM.numpy()\n",
    "            SM_max = np.abs(SM).max()#.item()\n",
    "            SM_min = -SM_max\n",
    "\n",
    "        if figsize is None: figsize = (self.opt.fig_width, self.opt.fig_width/self.opt.phi)\n",
    "\n",
    "        fig, axs = plt.subplots(1, N_SM_show, figsize=figsize, subplotpars=subplotpars)\n",
    "        for i_SM in range(N_SM_show):\n",
    "            ax = axs[i_SM]\n",
    "            ax.set_axisbelow(True)\n",
    "\n",
    "            ax.pcolormesh(SM[i_SM, :, :], cmap=cmap, vmin=SM_min, vmax=SM_max)\n",
    "            #ax.imshow(SM[:, i_SM, :], cmap=cmap, vmin=SM_min, vmax=SM_max, interpolation='none')\n",
    "            \n",
    "            ax.set_xlim(0, SM.shape[2])\n",
    "\n",
    "            ax.set_xlabel('Delay')\n",
    "            # ax.set_title(f'Motif {i_SM+1}', color='k' if colors is None else colors[i_SM])\n",
    "            # ax.text(-25, -.2*self.opt.N_pre, f'#{i_SM+1}', color='k' if colors is None else colors[i_SM])\n",
    "            t = ax.text(.40*self.opt.N_SM_time, .95*self.opt.N_pre, f'SM #{i_SM+1}', color='k' if colors is None else colors[i_SM])\n",
    "            t.set_bbox(dict(facecolor='white', edgecolor='white'))\n",
    "            if not aspect is None: ax.set_aspect(aspect)\n",
    "\n",
    "            ax.set_ylim(0, self.opt.N_pre)\n",
    "            ax.set_yticks(np.arange(0, self.opt.N_pre, 1)+.5)\n",
    "            if i_SM>0: \n",
    "                ax.set_yticklabels([])\n",
    "            else:\n",
    "                ax.set_yticklabels(np.arange(0, self.opt.N_pre, 1)+1)\n",
    "\n",
    "            for side in ['top', 'right']: ax.spines[side].set_visible(False)\n",
    "            #ax.set_xticks([0, self.opt.N_SM_time//2, self.opt.N_SM_time-1])\n",
    "            ax.set_xticks([1, self.opt.N_SM_time//3, (self.opt.N_SM_time*2)//3, self.opt.N_SM_time-1])\n",
    "            # ax.set_xticklabels([0, (self.opt.N_SM_time//2), (self.opt.N_SM_time)])\n",
    "            ax.xaxis.set_minor_locator(matplotlib.ticker.AutoMinorLocator(self.opt.N_SM_time//4))\n",
    "            #ax.xaxis.set_minor_locator(AutoMinorLocator(4))\n",
    "            #ax.set_xticklabels([-(self.opt.N_SM_time//2), 0, self.opt.N_SM_time//2])\n",
    "            # ax.grid(True, axis='y', linestyle='-', lw=1)\n",
    "            # ax.grid(True, axis='x', which='both', linestyle='-', lw=.1)\n",
    "\n",
    "        axs[0].set_ylabel('@ Neuron')\n",
    "        return fig, axs\n",
    "\n",
    "    def plot_a_histo(self, a,  xticks=1, spikelength=.9, colors=None, figsize=None, subplotpars=subplotpars):\n",
    "        if figsize is None: figsize = (self.opt.fig_width, self.opt.fig_width/self.opt.phi)\n",
    "\n",
    "        fig, ax = plt.subplots(1, 1, figsize=figsize, subplotpars=subplotpars)\n",
    "        ax.step(np.arange(self.opt.N_pre), a.numpy().mean(axis=(0, 2)), label='firing rate')\n",
    "        ax.hlines(self.opt.p_A, 0, self.opt.N_pre, linestyles='--', color='orange', label='probability of firing for each address')\n",
    "        ax.set_xlabel('address')\n",
    "        ax.set_ylim(0)\n",
    "        ax.legend()\n",
    "        return fig, ax\n",
    "\n",
    "    def plot_inference(self, b, b_hat, b_hat_bin, i_trial = 0, t_min = 100, t_max = 150, figsize=None, subplotpars=subplotpars):\n",
    "        if figsize is None: figsize = (self.opt.fig_width, self.opt.fig_width/self.opt.phi)\n",
    "        fig, ax = plt.subplots(1, 3, figsize=figsize, subplotpars=subplotpars)\n",
    "        ax[2].imshow(b_hat_bin[i_trial, :, t_min:t_max])\n",
    "        ax[0].imshow(b[i_trial, :, t_min:t_max])\n",
    "        ax[1].imshow(b_hat[i_trial, :, t_min:t_max])\n",
    "        return fig, ax\n",
    "\n",
    "    def plot_inference_histo(self, b_hat, nb_bins = 100, figsize=None, subplotpars=subplotpars):\n",
    "        if figsize is None: figsize = (self.opt.fig_width, self.opt.fig_width/self.opt.phi)\n",
    "\n",
    "        fig, ax = plt.subplots(1, 1, figsize=figsize, subplotpars=subplotpars)\n",
    "        ax.hist(b_hat.numpy().ravel(), bins=nb_bins)\n",
    "        ax.set_ylabel('smarts')\n",
    "        ax.set_xlabel('value of b_hat')\n",
    "        ax.set_yscale('log')\n",
    "        return fig, ax\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "1f0ab6b7-7f43-4c56-80e3-bf57b3805931",
   "metadata": {},
   "outputs": [],
   "source": [
    "env = ABCD(opt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "a607b61d-87de-426e-aa45-e4f314a28ef0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Params(datetag='2023-09-27', N_pre=128, N_SM_time=31, N_SMs=233, E_SM=4.0, p_SM=0.05, T_trial=1.0, tau_decay=0.4, tau_rise=0.05, N_time=1024, N_trials=10, p_B=0.001, p_A=0.001, seed=42, verbose=False, fig_width=12, phi=1.61803, N_SM_show=5)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env.opt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fc7e8553-cf29-4c78-a585-8e669a65b84e",
   "metadata": {},
   "source": [
    "#### dev sandox"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "45cf7c59",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(116, torch.Size([233, 233, 3]))"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(env.spike)//2, env.spike.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "e70f21d5",
   "metadata": {},
   "outputs": [],
   "source": [
    "env = ABCD(opt)\n",
    "a, b, SM = env.generative_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "2f4b4904",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([233, 233, 3])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env.spike.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "54160c9c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor(0.), tensor(1.))"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.min(), a.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "438b3711",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([128, 233, 31])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env.temporal_mod.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f65f41d0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([233, 128, 31])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "SM.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "72d580f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x2adfaed90>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "vmax = SM[0, :, :].abs().max()\n",
    "plt.imshow(SM[0, :, :], cmap='seismic', vmin=-vmax, vmax=vmax)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "c5f59873-584d-43ad-a1e5-84172faee5e5",
   "metadata": {
    "tags": []
   },
   "source": [
    "# generative model for spiking motifs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "3466fd28-1975-45d1-ad10-7d604318bb3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "opt = Params()\n",
    "env = ABCD(opt)\n",
    "a, b, SM = env.generative_model()\n",
    "# env.plot_SM(SM);\n",
    "# TODO find a better rule to generate SMs"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "4cb1eb03-3ee1-4599-bb89-5d769ce9d872",
   "metadata": {},
   "source": [
    "Draw one instance of single SMs:"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "22abf88b-b2f9-4b16-af19-b7ca6aba7035",
   "metadata": {},
   "source": [
    "Draw the occurrences of SMs:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "09b07bf1-68db-46e3-b3ed-bc41f2aec49f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x741.643 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "env.plot_b(b);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13bed7c5-824c-44e2-ad33-1508ebda8106",
   "metadata": {},
   "source": [
    "Resulting raster plot:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "c4b278db-9cca-4d21-8428-170e1b61daf6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x741.643 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "env.plot_a(a, i_trial=7);\n",
    "# TODO correct this firing rate"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2fb8a4d4",
   "metadata": {},
   "source": [
    "TODO: circular padding?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "9c31bb02",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor(0.020),\n",
       " tensor(0.001),\n",
       " tensor([0.001, 0.002, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001,\n",
       "         0.001]))"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.mean(), b.mean(), b.mean(axis=(0, -1))[:20]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "23fc3f4f-9180-4c2d-ad50-da05e3b35064",
   "metadata": {},
   "source": [
    "# inference model for spiking motifs\n",
    "\n",
    "knowing the raster plot `a`, let's try to infer `b` :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "4a58a7a8-0aa2-46c9-aa18-c843eb415d64",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<Figure size 1200x741.643 with 1 Axes>,\n",
       " <Axes: xlabel='value of b_hat', ylabel='smarts'>)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x741.643 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "b_hat, b_hat_bin = env.inference_with_SMs(a, b, SM)\n",
    "env.plot_inference_histo(b_hat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "684bb4ac-a02c-4bda-9227-7ef2796d4462",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<Figure size 1200x741.643 with 3 Axes>,\n",
       " array([<Axes: >, <Axes: >, <Axes: >], dtype=object))"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x741.643 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "env.plot_inference(b, b_hat, b_hat_bin)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5b0f0d56-ec2e-4702-899c-9694cfed9eaa",
   "metadata": {},
   "source": [
    "## test detection model, knowing the SMs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "5d083bed-071c-4ec5-bc52-af18108c13b2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "386 ms ± 3.22 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "accuracy, TP, TN = env.test_model(SM, SM)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "653935b2-4ad1-4a6b-8bff-d5bb0279af8a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy=99.99%, TP=95.02%, TN=100.00%\n"
     ]
    }
   ],
   "source": [
    "accuracy, TP, TN = env.test_model(SM, SM)\n",
    "print(f'accuracy={accuracy.item()*100:.2f}%, TP={TP.item()*100:.2f}%, TN={TN.item()*100:.2f}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "05aac9f2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "387 ms ± 1.73 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)\n"
     ]
    }
   ],
   "source": [
    "%%timeit\n",
    "accuracy, TP, TN = env.test_model(SM, SM, method='corr')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "cf183d91",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy=99.96%, TP=77.75%, TN=99.98%\n"
     ]
    }
   ],
   "source": [
    "accuracy, TP, TN = env.test_model(SM, SM, method='corr')\n",
    "print(f'accuracy={accuracy.item()*100:.2f}%, TP={TP.item()*100:.2f}%, TN={TN.item()*100:.2f}%')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "1176a73f-9aaf-4581-aff1-fef7d8ca5568",
   "metadata": {},
   "source": [
    "## Figure toy model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "150f9909-f630-4d1e-b81a-e8df61e4fb86",
   "metadata": {},
   "outputs": [],
   "source": [
    "N_SMs = 4\n",
    "opt_fig_toy = Params(\n",
    "    datetag = datetag + '_fig1',\n",
    "    N_pre = 10, \n",
    "    N_SMs = N_SMs,\n",
    "    N_SM_time = 31,\n",
    "    E_SM = 100.,\n",
    "    p_SM = .02,\n",
    "    p_A = 0.001,\n",
    "    p_B = 0.001, # not relevant here as we select B below\n",
    "    N_SM_show = N_SMs,\n",
    "    N_trials = 1,\n",
    "    N_time = 240,\n",
    "    seed = 1973,\n",
    "    fig_width = 3.25,\n",
    "    )\n",
    "                    \n",
    "env_fig1 = ABCD(opt_fig_toy)\n",
    "# a, b, SM = env_fig1.generative_model()\n",
    "SM = env_fig1.set_SM(seed=opt_fig_toy.seed)\n",
    "b = env_fig1.get_b(seed=opt_fig_toy.seed+1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "e4ebc802-70e0-4f34-9965-bd96a2ee629f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Params(datetag='2023-09-27_fig1', N_pre=10, N_SM_time=31, N_SMs=4, E_SM=100.0, p_SM=0.02, T_trial=1.0, tau_decay=0.4, tau_rise=0.05, N_time=240, N_trials=1, p_B=0.001, p_A=0.001, seed=1973, verbose=False, fig_width=3.25, phi=1.61803, N_SM_show=4)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env_fig1.opt\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "fd2a317e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10, 4, 31)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env_fig1.opt.N_pre, env_fig1.opt.N_SMs, env_fig1.opt.N_SM_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "d85f4976-f188-4fb5-87da-f133ee654fa3",
   "metadata": {},
   "outputs": [],
   "source": [
    "subplotpars_fig1 = matplotlib.figure.SubplotParams(left=0., right=1., bottom=0., top=1., wspace=0.20, hspace=0.0,)\n",
    "colors = ['r', 'g' , 'b' , 'm']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "1d0f5ab4-ad12-4570-bc5e-b43fa9ab5950",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([1, 4, 240])"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b = torch.zeros(b.shape)\n",
    "b[0, 0,  20] = True\n",
    "b[0, 0,  89] = True\n",
    "b[0, 1, 70] = True\n",
    "b[0, 1, 104] = True\n",
    "b[0, 1, 160] = True\n",
    "b[0, 1, 197] = True\n",
    "b[0, 2,  42] = True\n",
    "b[0, 2, 130] = True\n",
    "b[0, 3, 160] = True\n",
    "b.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "0ab69510",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([4, 10, 31])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "SM.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "d57b892f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor(0.004), tensor(221.585), tensor(-177.268), tensor(221.585))"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "SM.mean(), SM.abs().max(), SM.min(), SM.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "27d211ce-660c-4465-9856-dfa6b856f136",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x200.862 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "shift = -0.14\n",
    "fig, axs = env_fig1.plot_SM(SM, colors=colors, subplotpars=subplotpars_fig1)\n",
    "fig.text(shift, 0.95,'(b)', horizontalalignment='center', verticalalignment='center', weight='bold')\n",
    "if not figpath is None: printfig(fig, 'THC_toy-b')\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "c45e47f9-3924-4bc5-8aca-588d4c8bbb29",
   "metadata": {},
   "source": [
    "Then from the following raster plot, each SM described if the figure above will be selective to the spatio-temporal pattern of the same color: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "b2eb570d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x200.862 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = env_fig1.plot_b(b, colors=colors, yticks=env_fig1.opt.N_SMs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "1beb8304",
   "metadata": {},
   "outputs": [],
   "source": [
    "a = env_fig1.draw_a(b, SM)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "0b90dc53",
   "metadata": {},
   "outputs": [],
   "source": [
    "b_hat, b_hat_bin = env_fig1.inference_with_SMs(a, b, SM)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "15186ba5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(torch.Size([240]),\n",
       " torch.Size([1, 10, 240]),\n",
       " torch.Size([1, 4, 240]),\n",
       " torch.Size([1, 4, 240]))"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env_fig1.time.shape, a.shape, b_hat.shape, b_hat_bin.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62f3eb32",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "6b9b9bab-3c77-4aad-9fb5-59a52b842112",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x200.862 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = env_fig1.plot_b(b_hat_bin, colors=colors, subplotpars=subplotpars_fig1, yticks=env_fig1.opt.N_SMs)\n",
    "fig.text(shift/2, 0.95,'(c)', horizontalalignment='center', verticalalignment='center', weight='bold')\n",
    "b_hat_np = b_hat.squeeze(0).numpy()\n",
    "b_hat_np /= np.abs(b_hat_np).max() / .8\n",
    "for i_SM in range(env_fig1.opt.N_SMs):\n",
    "    ax.plot(b_hat_np[i_SM, :]+i_SM+.5, color=colors[i_SM], alpha=.35)\n",
    "    b_star = np.where(b[0, i_SM, :]==1)[0]\n",
    "    ax.scatter(b_star, (i_SM+.85)*np.ones_like(b_star), color=colors[i_SM], marker='*')\n",
    "if not figpath is None: printfig(fig, 'THC_toy-c')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "78122510-720c-4a35-b921-b01f75ccc465",
   "metadata": {},
   "source": [
    "It results in the following activation of the postsynaptic neurons (activation probability with a Bernoullin trial):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "f1741d0b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([160])"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.where(b[0, i_SM, :]==1)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "20b1de1e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10, torch.Size([1, 10, 240]), torch.Size([1, 4, 240]))"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env_fig1.opt.N_pre, a.shape, b.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "76885744-43b7-4067-84cd-bbcb464d0ab7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x200.862 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x200.862 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for suff in ['_k', '']:\n",
    "    fig, ax = env_fig1.plot_a(a, b, SM, shift=env_fig1.opt.N_SM_time//4, # HACK\n",
    "                              colors=colors if suff=='' else ['k'], subplotpars=subplotpars_fig1, yticks=env_fig1.opt.N_pre)\n",
    "    \n",
    "    if not figpath is None: \n",
    "        fig.text(shift, 0.95, '(a)' if suff=='_k' else '(d)', horizontalalignment='center', verticalalignment='center', weight='bold')\n",
    "        printfig(fig, 'THC_toy-a' + suff)\n",
    "    else:\n",
    "        fig.text(0.05, 0.95,'(before)' if suff=='_k' else '(after)', horizontalalignment='center', verticalalignment='center', weight='bold')\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "d2d19699-7afd-44b6-af46-d21d84712b4d",
   "metadata": {},
   "source": [
    "TODO: montrer des SMs générant des vagues, du bruit de Poisson, des one-spike per neuron, ..."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3edc511f",
   "metadata": {},
   "source": [
    "## testing inference"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "36e64789",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Params(datetag='2023-09-27', N_pre=128, N_SM_time=31, N_SMs=233, E_SM=4.0, p_SM=0.05, T_trial=1.0, tau_decay=0.4, tau_rise=0.05, N_time=1024, N_trials=10, p_B=0.001, p_A=0.001, seed=42, verbose=False, fig_width=12, phi=1.61803, N_SM_show=5)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env = ABCD(opt)\n",
    "env.opt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "689a2727",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy=99.99%, TP=95.06%, TN=100.00%\n"
     ]
    }
   ],
   "source": [
    "accuracy, TP, TN = env.test_model(SM=env.set_SM(seed=42), SM_true=env.set_SM(seed=42))\n",
    "print(f'accuracy={accuracy.item()*100:.2f}%, TP={TP.item()*100:.2f}%, TN={TN.item()*100:.2f}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "60f646a3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy=99.96%, TP=77.15%, TN=99.98%\n"
     ]
    }
   ],
   "source": [
    "accuracy, TP, TN = env.test_model(SM=env.set_SM(seed=42), SM_true=env.set_SM(seed=42), method='corr')\n",
    "print(f'accuracy={accuracy.item()*100:.2f}%, TP={TP.item()*100:.2f}%, TN={TN.item()*100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "158b51df-c7c8-4981-84f9-00d742fee70b",
   "metadata": {
    "tags": []
   },
   "source": [
    "## scanning parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "8850581b-7e7e-4824-8470-b7b8067739cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_cache = 'data_cache'\n",
    "%mkdir -p {data_cache}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "5b2b9e6e-b1e5-49f2-9a46-79694c7884da",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2023-09-27_scan'"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "opt_scan = Params( datetag = datetag + '_scan',\n",
    "                    N_time = 2**14,\n",
    "                    fig_width = 3.25,\n",
    "                    )\n",
    "opt_scan.datetag"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "435b4305",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16384"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "opt_scan.N_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "c9db5180-60e7-4b6f-84e3-483d09d15b52",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data_cache/2023-09-27_scan_E_SM.json\n",
      "data_cache/2023-09-27_scan_N_SM_time.json\n",
      "data_cache/2023-09-27_scan_N_SMs.json\n",
      "data_cache/2023-09-27_scan_N_pre.json\n",
      "data_cache/2023-09-27_scan_p_A.json\n",
      "data_cache/2023-09-27_scan_p_B.json\n",
      "data_cache/2023-09-27_scan_p_SM.json\n",
      "data_cache/2023-09-27_scan_tau_decay.json\n",
      "data_cache/2023-09-27_scan_tau_rise.json\n"
     ]
    }
   ],
   "source": [
    "%ls {data_cache}/{opt_scan.datetag}_*.json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "b3882b67-a689-4e2c-ab98-7a446c5d5ccc",
   "metadata": {},
   "outputs": [],
   "source": [
    "# %rm -f {data_cache}/{datetag}_scan_*.json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "065a2ebf-4dd6-48e0-a885-c527fe358c96",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "filename='data_cache/2023-09-27_scan_N_SMs.json'\n",
      "     N_SMs  accuracy        TP        TN  method      time\n",
      "0        8  0.999951  0.976744  0.999976  HD-SNN  0.063379\n",
      "1        8  0.999951  0.974026  0.999976  HD-SNN  0.052444\n",
      "2        8  0.999951  0.970149  0.999976  HD-SNN  0.048137\n",
      "3        8  0.999829  0.919540  0.999914  HD-SNN  0.050146\n",
      "4        8  0.999951  0.973684  0.999976  HD-SNN  0.049077\n",
      "..     ...       ...       ...       ...     ...       ...\n",
      "255   2351  0.998531  0.242215  0.999263    corr  7.554011\n",
      "256   2351  0.998544  0.244475  0.999275    corr  6.927373\n",
      "257   2351  0.998513  0.236915  0.999259    corr  6.687131\n",
      "258   2351  0.998533  0.243135  0.999269    corr  6.627032\n",
      "259   2351  0.998555  0.250875  0.999275    corr  7.680571\n",
      "\n",
      "[260 rows x 6 columns]\n",
      "filename='data_cache/2023-09-27_scan_N_SM_time.json'\n",
      "     N_SM_time  accuracy        TP        TN  method      time\n",
      "0            1  1.000000       NaN  1.000000  HD-SNN  0.069434\n",
      "1            1  1.000000       NaN  1.000000  HD-SNN  0.063933\n",
      "2            1  1.000000       NaN  1.000000  HD-SNN  0.059802\n",
      "3            1  1.000000       NaN  1.000000  HD-SNN  0.060135\n",
      "4            1  1.000000       NaN  1.000000  HD-SNN  0.054192\n",
      "..         ...       ...       ...       ...     ...       ...\n",
      "135        127  0.999940  0.966430  0.999970    corr  0.859307\n",
      "136        127  0.999954  0.973301  0.999977    corr  0.850763\n",
      "137        127  0.999951  0.972048  0.999976    corr  0.839471\n",
      "138        127  0.999953  0.973746  0.999977    corr  0.843567\n",
      "139        127  0.999951  0.972443  0.999975    corr  0.892726\n",
      "\n",
      "[140 rows x 6 columns]\n",
      "filename='data_cache/2023-09-27_scan_N_pre.json'\n",
      "     N_pre  accuracy        TP        TN  method      time\n",
      "0        8  0.998348  0.133626  0.999173  HD-SNN  0.150681\n",
      "1        8  0.998264  0.128367  0.999131  HD-SNN  0.154906\n",
      "2        8  0.998317  0.116146  0.999158  HD-SNN  0.142787\n",
      "3        8  0.998338  0.127912  0.999169  HD-SNN  0.137957\n",
      "4        8  0.998292  0.146923  0.999145  HD-SNN  0.156435\n",
      "..     ...       ...       ...       ...     ...       ...\n",
      "255   2351  0.999998  0.999143  0.999999    corr  2.499372\n",
      "256   2351  0.999999  0.999553  1.000000    corr  2.534012\n",
      "257   2351  1.000000  1.000000  1.000000    corr  2.531489\n",
      "258   2351  0.999997  0.998680  0.999999    corr  2.543407\n",
      "259   2351  0.999998  0.999138  0.999999    corr  2.536328\n",
      "\n",
      "[260 rows x 6 columns]\n",
      "filename='data_cache/2023-09-27_scan_E_SM.json'\n",
      "     E_SM  accuracy        TP        TN  method      time\n",
      "0       1  0.998129  0.018901  0.999064  HD-SNN  0.334534\n",
      "1       1  0.998049  0.020202  0.999023  HD-SNN  0.328397\n",
      "2       1  0.998132  0.019358  0.999065  HD-SNN  0.336163\n",
      "3       1  0.998114  0.010989  0.999056  HD-SNN  0.337661\n",
      "4       1  0.998029  0.015906  0.999014  HD-SNN  0.348119\n",
      "..    ...       ...       ...       ...     ...       ...\n",
      "135   127  1.000000  1.000000  1.000000    corr  0.442675\n",
      "136   127  1.000000  1.000000  1.000000    corr  0.449080\n",
      "137   127  0.999999  0.999569  1.000000    corr  0.448772\n",
      "138   127  0.999999  0.999574  1.000000    corr  0.441505\n",
      "139   127  0.999999  0.999572  1.000000    corr  0.447610\n",
      "\n",
      "[140 rows x 6 columns]\n",
      "filename='data_cache/2023-09-27_scan_tau_decay.json'\n",
      "     tau_decay  accuracy        TP        TN  method      time\n",
      "0          0.1  0.998340  0.129670  0.999169  HD-SNN  0.378289\n",
      "1          0.1  0.998241  0.116582  0.999119  HD-SNN  0.385068\n",
      "2          0.1  0.998360  0.138583  0.999180  HD-SNN  0.385842\n",
      "3          0.1  0.998337  0.127912  0.999168  HD-SNN  0.381018\n",
      "4          0.1  0.998289  0.145668  0.999144  HD-SNN  0.381509\n",
      "..         ...       ...       ...       ...     ...       ...\n",
      "175        0.9  0.999955  0.976684  0.999977    corr  0.443320\n",
      "176        0.9  0.999947  0.972173  0.999974    corr  0.442818\n",
      "177        0.9  0.999941  0.970289  0.999971    corr  0.445000\n",
      "178        0.9  0.999975  0.986882  0.999987    corr  0.445637\n",
      "179        0.9  0.999962  0.980511  0.999981    corr  0.442980\n",
      "\n",
      "[180 rows x 6 columns]\n",
      "filename='data_cache/2023-09-27_scan_tau_rise.json'\n",
      "     tau_rise  accuracy        TP        TN  method      time\n",
      "0        0.01  0.999975  0.986813  0.999987  HD-SNN  0.443718\n",
      "1        0.01  0.999967  0.983586  0.999984  HD-SNN  0.442027\n",
      "2        0.01  0.999978  0.988561  0.999989  HD-SNN  0.442102\n",
      "3        0.01  0.999982  0.990769  0.999991  HD-SNN  0.445676\n",
      "4        0.01  0.999974  0.987024  0.999987  HD-SNN  0.445614\n",
      "..        ...       ...       ...       ...     ...       ...\n",
      "115      0.90  0.998065  0.005172  0.999032    corr  0.318829\n",
      "116      0.90  0.998022  0.007569  0.999010    corr  0.322723\n",
      "117      0.90  0.998081  0.005215  0.999040    corr  0.309788\n",
      "118      0.90  0.998093  0.009146  0.999046    corr  0.308627\n",
      "119      0.90  0.998077  0.008643  0.999038    corr  0.316173\n",
      "\n",
      "[120 rows x 6 columns]\n",
      "filename='data_cache/2023-09-27_scan_p_SM.json'\n",
      "      p_SM  accuracy        TP        TN  method      time\n",
      "0    0.005  0.999385  0.677363  0.999692  HD-SNN  0.249565\n",
      "1    0.005  0.999379  0.688131  0.999689  HD-SNN  0.247990\n",
      "2    0.005  0.999416  0.693357  0.999708  HD-SNN  0.258380\n",
      "3    0.005  0.999425  0.698022  0.999712  HD-SNN  0.256826\n",
      "4    0.005  0.999392  0.696526  0.999696  HD-SNN  0.259028\n",
      "..     ...       ...       ...       ...     ...       ...\n",
      "235  1.000  0.999815  0.904987  0.999907    corr  0.440625\n",
      "236  1.000  0.999776  0.883406  0.999888    corr  0.447725\n",
      "237  1.000  0.999769  0.881647  0.999884    corr  0.442866\n",
      "238  1.000  0.999800  0.893966  0.999900    corr  0.447613\n",
      "239  1.000  0.999764  0.877338  0.999882    corr  0.446120\n",
      "\n",
      "[240 rows x 6 columns]\n",
      "filename='data_cache/2023-09-27_scan_p_A.json'\n",
      "       p_A  accuracy        TP        TN  method      time\n",
      "0    0.001  0.999881  0.937582  0.999940  HD-SNN  0.446202\n",
      "1    0.001  0.999896  0.947811  0.999948  HD-SNN  0.453300\n",
      "2    0.001  0.999908  0.951606  0.999954  HD-SNN  0.445786\n",
      "3    0.001  0.999910  0.952967  0.999955  HD-SNN  0.448156\n",
      "4    0.001  0.999912  0.956049  0.999956  HD-SNN  0.446265\n",
      "..     ...       ...       ...       ...     ...       ...\n",
      "155  0.500  0.998546  0.253230  0.999273    corr  0.437053\n",
      "156  0.500  0.998778  0.371552  0.999388    corr  0.441794\n",
      "157  0.500  0.998800  0.392706  0.999399    corr  0.440919\n",
      "158  0.500  0.998886  0.427895  0.999442    corr  0.440956\n",
      "159  0.500  0.998717  0.323607  0.999358    corr  0.440026\n",
      "\n",
      "[160 rows x 6 columns]\n",
      "filename='data_cache/2023-09-27_scan_p_B.json'\n",
      "       p_B  accuracy        TP        TN  method      time\n",
      "0    0.001  0.999881  0.937582  0.999940  HD-SNN  0.442927\n",
      "1    0.001  0.999896  0.947811  0.999948  HD-SNN  0.441032\n",
      "2    0.001  0.999908  0.951606  0.999954  HD-SNN  0.447361\n",
      "3    0.001  0.999910  0.952967  0.999955  HD-SNN  0.443855\n",
      "4    0.001  0.999912  0.956049  0.999956  HD-SNN  0.452339\n",
      "..     ...       ...       ...       ...     ...       ...\n",
      "155  0.500  0.606247  0.594893  0.616983    corr  0.460322\n",
      "156  0.500  0.605156  0.593365  0.616283    corr  0.447884\n",
      "157  0.500  0.606234  0.594219  0.617558    corr  0.453909\n",
      "158  0.500  0.603978  0.591834  0.615421    corr  0.450483\n",
      "159  0.500  0.605717  0.593583  0.617148    corr  0.449912\n",
      "\n",
      "[160 rows x 6 columns]\n"
     ]
    }
   ],
   "source": [
    "scan_dicts= {'N_SMs' : [8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987, 1364, 2351],\n",
    "             'N_SM_time' : [1, 3, 7, 15, 31, 63, 127],\n",
    "             'N_pre' : [8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987, 1364, 2351],\n",
    "             'E_SM' : [1, 3, 7, 15, 31, 63, 127],\n",
    "             'tau_decay':[0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9],\n",
    "             'tau_rise':[0.01,0.05,0.1,0.2,0.5,0.9],\n",
    "             # TODO : use geomspace\n",
    "             'p_SM' : [k/1000 for k in [5, 10, 15, 25, 40, 65, 105, 170, 275, 445, 720, 1000]],\n",
    "             'p_A':[0.001,0.002,0.005,0.01,0.02,0.05,0.1,0.5],\n",
    "             'p_B':[0.001,0.002,0.005,0.01,0.02,0.05,0.1,0.5]             \n",
    "\n",
    "            }\n",
    "\n",
    "label_dicts= {'N_SMs' : 'number of motifs',\n",
    "             'N_SM_time' : 'duration of motifs',\n",
    "             'N_pre' : 'number of presynaptic inputs',\n",
    "             'E_SM' : 'peak excitability of motifs',\n",
    "             'p_SM' : 'sparsity of motifs',\n",
    "             'tau_decay': 'time constant of decay',\n",
    "             'tau_rise': 'time constant of rise',\n",
    "             'p_A': 'apriori probability of firing for each presynaptic address',\n",
    "             'p_B':  'apriori probability of firing for each motif',\n",
    "            }\n",
    "import os\n",
    "import time\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "N_cv = 10\n",
    "seed = 1000 # np.random.randint(1000)\n",
    "verb = False\n",
    "\n",
    "for key in scan_dicts:\n",
    "    filename = f'{data_cache}/{opt_scan.datetag}_{key}.json'\n",
    "    print(f'{filename=}')\n",
    "    if os.path.isfile(filename):\n",
    "        df_scan = pd.read_json(filename)\n",
    "    else:\n",
    "        measure_columns = [key, 'accuracy', 'TP', 'TN', 'method', 'time']\n",
    "        df_scan = pd.DataFrame([], columns=measure_columns)\n",
    "        i_loc = 0\n",
    "        for method in ['HD-SNN', 'corr']: \n",
    "            for i_value, value in enumerate(scan_dicts[key]):\n",
    "                if verb: print('i_value', i_value + 1, ' /', len(scan_dicts[key]), end='\\t')\n",
    "\n",
    "                for i_cv in  range(N_cv):\n",
    "                    # tic\n",
    "                    since = time.time()\n",
    "                    new_dict = asdict(opt)\n",
    "                    # print(key, value)\n",
    "                    new_dict[key] = value\n",
    "                    new_opt = Params(**new_dict)\n",
    "                    # print(new_opt)\n",
    "                    # make a new model\n",
    "                    new_env = ABCD(new_opt)\n",
    "                    # draw a new set of SMs\n",
    "                    SM = new_env.set_SM(seed=seed+i_loc)\n",
    "                    # test it\n",
    "                    accuracy, TP, TN = new_env.test_model(SM, SM, seed=seed+i_loc, method=method)\n",
    "                    # toc\n",
    "                    elapsed_time = time.time() - since\n",
    "                    df_scan.loc[i_loc] = {key:value, 'accuracy':accuracy.item(), 'TP':TP.item(), 'TN':TN.item(), 'method':method, 'time':elapsed_time}\n",
    "                    i_loc += 1\n",
    "                    if verb: print(f\" - One pass completed in {elapsed_time // 60:.0f}m {elapsed_time % 60:.0f}s\")\n",
    "\n",
    "        df_scan.to_json(filename)\n",
    "    print(df_scan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "0f7257b4-f640-4500-9cb5-ea68e11ad96e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading filename='data_cache/2023-09-27_scan_N_SMs.json'...\n",
      "     N_SMs  accuracy        TP        TN  method      time\n",
      "0        8  0.999951  0.976744  0.999976  HD-SNN  0.063379\n",
      "1        8  0.999951  0.974026  0.999976  HD-SNN  0.052444\n",
      "2        8  0.999951  0.970149  0.999976  HD-SNN  0.048137\n",
      "3        8  0.999829  0.919540  0.999914  HD-SNN  0.050146\n",
      "4        8  0.999951  0.973684  0.999976  HD-SNN  0.049077\n",
      "..     ...       ...       ...       ...     ...       ...\n",
      "255   2351  0.998531  0.242215  0.999263    corr  7.554011\n",
      "256   2351  0.998544  0.244475  0.999275    corr  6.927373\n",
      "257   2351  0.998513  0.236915  0.999259    corr  6.687131\n",
      "258   2351  0.998533  0.243135  0.999269    corr  6.627032\n",
      "259   2351  0.998555  0.250875  0.999275    corr  7.680571\n",
      "\n",
      "[260 rows x 6 columns]\n",
      "Bbox(x0=0.125, y0=0.25, x1=0.95, y1=0.975)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVcAAAFHCAYAAAAV5w+DAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAABBX0lEQVR4nO3dd1hT1xsH8G+YgooDFAKiuKCoKAKCOH51llrraKt1VRStYh2g1FaxKlpbcYtVKmCddeGorRaqVSpOqlbEUQW3KA3gKDJNIHl/f1iiYSaQK0Tfz/PwSG7uPTnXS745Oefce0VERGCMMaZVelVdAcYYex1xuDLGmAA4XBljTAAcrowxJgAOV8YYEwCHK2OMCYDDlTHGBMDhyhhjAuBwZYwxAXC4MsaYAKo0XI8fP45+/frB2toaIpEIP//8c7nbxMbGwsXFBcbGxmjRogU2bdokeD0ZY0xTVRquOTk5aNeuHUJDQ9Va/86dO+jbty+6d++OhIQETJ06FZ9++ikOHTokcE0ZY0wzoupy4RaRSIR9+/Zh4MCBpa4zY8YMREVF4cqVK8plQ4cORUZGBg4ePPgKaskYY+oxqOoKaCIuLg69evVSWebl5YWpU6eWuo1UKoVUKlU+VigUePLkCczNzSESiYSqKmPsNUBEyMrKgrW1NfT0NPuir1PhmpqaCktLS5VllpaWyMzMRF5eHkxMTIptExwcjPnz57+qKjLGXkP3799Ho0aNNNpGp8K1IgIDAxEQEKB8/PTpUzRu3Bj379+HmZmZcnlwcDAWLVqkdrkzZ85EYGCgVuvKGKteMjMzYWtri9q1a2u8rU6Fq5WVFdLS0lSWpaWlwczMrMRWKwAYGxvD2Ni42HIzMzOVcPXz88PgwYOVj/Py8tClSxcAwMmTJ4uVLxaLVbYvSiKRQCKRlL9TL5UnFovVXp8x9upUpAtRp8LV09MT0dHRKssOHz4MT0/PSpddNNxycnKUvzs7O6NmzZoalRceHq5Rd0RQUBDmzZun0WswxqqvKg3X7Oxs3Lx5U/n4zp07SEhIQP369dG4cWMEBgYiJSUFW7ZsAQBMmDABa9aswZdffokxY8bgjz/+wK5duxAVFVVVu1AqX19f9O/fX/lYnZZwebg1zJjuqNJw/euvv9C9e3fl48K+0VGjRmHTpk2QSCRITk5WPt+0aVNERUVh2rRpWLVqFRo1aoQffvgBXl5ela5L0eDKy8tT/p6QkFBiGJYVXNpuCQPcGmZMl1Sbea6vSmZmJurUqYOnT5+q9JnOmzdP68H18tMyWQ6Cg2sBAAIDs2FkVLPE9cpS0gdAea1hbrkyVnGl5YU6dKrPVUhZWb4YP/7F1/jc3CfYurU3AOCTTw7D1LS+8jlfX/W+xmubtlvD3M3AmHA4XP9Tu7YYtWu/CI7s7HTl75aWbVGrVkPlYxeX8ssrGlz5+S+6GVJTE2Bo+KKVGR9fNcElRDeDEIHNHwJMF3G4CiQ8PBwRESUH18aNXVQeR0So3z+q2tXw4vdvvwWMjEperzS+vr6QSF601vPz85R18/E5qfIB8Hz98gNLiMDWhb5mXfgA0IU6vk44XAUycOBAXL7cUq11P/wQaN26tcA1ejU0CWx1u1e0PfNCiJDRhW8BuvAh9TrhcP1PVpYE2dkv/pBzc58of09Lu4SsrBd9rup8jf/555/x00/q/SH/9NPzP2RnZ+cy19N2V4MmrWsAEIur5s0mFosRHv5iP2SyF33NUVHOGg8OChEyRT9M8/OlOHBgLACgX7/1MDR8cSKLuh+mH364DH/+uULtegYEBGD58uWvtI6sdByu/zl/PhzHjpX8hisc2Cqkztd4IVpw2u5qKFrH8qjbLSBEd4g2CREyZX2YFpZdSN0PU23ThTq+Tjhc/+Pq6gsHB+2FYdFW48utLSsr1daWOgNkz19X/TCsSB3Lo86q2q4joP0WuxAhI8R+d+o0HW3bjtBamULUkZWOw/U/RWcLaCMMtU2TMHyd6liVLXZ1Q0aI/S76N1nZMoWoIw+SlY7D9T/dYuepPM6TyxD83+9dT3wLE/2XhuKhui4TVlW22KvqQ0pXLFu2DCtWaK9f+HXC4apDNGklVNXcWSFwGDJdxOGqQ8r6elxUVQ0WsTfL9OnTkZ2tvX7h1wmHqw4RYgYCY5XB3ypKx+GqQ4SYgcBYdfI6DZBxuDLGqo3X6SwyDlfGWLWhyQkez9evvmeRcbgyxqoNTU7wAAAnp+p7FhmHK2Os2hDilOyqwuH6n8fSLDyWZSsfSxX5yt9vZqfCWM9Q+dgsPl6tjvSXT0zgkxIYK58Qp2RXFQ7X/xyQnMfme8dKfM4vYaPqAteIKutI12ZgazoyK5FU35FZxqobDtf/9BO7opO5g1rruq3zfS1CRpOTEoCqu+QgY7qIw/U/5sa1YW5cW611XV6TSaRC3ImAMfYch+sbTJOTEp6v/8qqxpjO06vqCjDG2OuIw5UxxgTA3QI6RCKR4HrWi9H9VztdDND2DITX6bKIjBXF4apDwsPDMT8+osTnqst0Mb4sImPPcbjqEF9fX1gfUK9VWFXTxYS4LCK3hpku4nDVIWKxGPZq31OpaqaLCXFZRG4NM13E4cqqPb5IONNFHK5vME0GyADAQSJRK7i0fU0Fvkg400UcrgIRYmRf2zQaIAPwT7i4yr5u80VwmK7hcBWILozsazJABgD9fH3LXUcXPlQYexU4XAWiCyP7mgyQFa5fHiE+VDiwmS7icBWILozsC0GIDxVd+BbAWFEcrkyrhPhQ0YVvAYwVxeHKqr039VsA021VHq6hoaFYunQpUlNT0a5dO6xevRru7u6lrh8SEoK1a9ciOTkZFhYWGDRoEIKDg1GjRo1XWGvGmK7Q9I4b2uqzr9JwjYyMREBAAMLCwuDh4YGQkBB4eXkhKSkJDRs2LLb+9u3bMXPmTGzYsAGdOnXC9evXMXr0aIhEIqxYsaIK9oAxVt0tW7ZMo3wICAjA8uXLK/26VRquK1aswLhx4+Dj4wMACAsLQ1RUFDZs2ICZM2cWW//06dPo3Lkzhg8fDgCws7PDsGHDcObMmVdab6bb+FoF7FWosnCVyWQ4f/48AgMDlcv09PTQq1cvxMXFlbhNp06dsHXrVpw9exbu7u64ffs2oqOjMXLkyFJfRyqVQiqVKh9nZmZqbyeYTuJrFbxZpk+fjuzsEWqtq83Tp6ssXB89egS5XA5LS0uV5ZaWlkhMTCxxm+HDh+PRo0fo0qULiAgFBQWYMGECZs2aVerrBAcHY/589W/Cx15/RWcfSBX5yild3zn7qMyb5dkHuk+Tbx7aHA/VqTsRxMbGYuHChfj+++8RHx+Pn376CVFRUViwYEGp2wQGBuLp06fKn/v377/CGjPG3lRV1nK1sLCAvr4+0tLSVJanpaXBysqqxG3mzJmDkSNH4tNPPwUAODk5IScnB+PHj8dXX30FPb3inxXGxsYwNjbW/g4wncUnJbBXocrC1cjICK6uroiJicHAgQMBAAqFAjExMZg8eXKJ2+Tm5hYLUH19fQAAEQlaX/b64JMS2KtQpbMFAgICMGrUKLi5ucHd3R0hISHIyclRzh7w9vaGjY0NgoOfXwOpX79+WLFiBdq3bw8PDw/cvHkTc+bMQb9+/ZQhy1h5hDgpgWcgsKKqNFyHDBmChw8fYu7cuUhNTYWzszMOHjyoHORKTk5WaanOnj0bIpEIs2fPRkpKCho0aIB+/frh22+/rapdYAwAz0BgxVX5GVqTJ08utRsgNjZW5bGBgQGCgoIQFBT0CmrGmPr4bgmsqCoPV8ZeB3y3BFYUhytjWsJ3S2Av06l5rowxpiu45cqYFvDdElhRHK6MaQGfmMCK4nBlTAv4xARWFIcrY1rAd0tgRfGAFmOMCYDDlTHGBMDhyhhjAuBwZYwxAXC4MsaYADhcGWNMADwVi7FqKitLguxsvkasruJwZayaOn8+HMeO8TVidRWHK2PVlKurLxwc+BqxuorDlbFqqt/5cJXHeXIZCq9SMPhGlMplDF1c5r2yejH18IAWY4wJgFuujFVTj6VZeCzLVj7myxjqFg5XxqqpA5Lz2HzvWInP8WUMqz8OV8aqqX5iV3Qyd1BrXb6MYfXD4cpYNWVuXBvmxrXVWpcvY1j98IAWY4wJgMOVMcYEwOHKGGMC4HBljDEBcLgyxpgAOFwZY0wAHK6MMSYADlfGGBMAhytjjAmAw5UxxgTAp78y9oaQSCSQSPi2Ma8Khytjb4jw8HBERPBtY14VDlfG3hC+vr6QSPi2Ma+KxuFqZ2eHMWPGYPTo0WjcuLEQdWKMCUAsFmNY0otbx/BtY4Sl8YDW1KlT8dNPP6FZs2bo3bs3du7cCalUWuEKhIaGws7ODjVq1ICHhwfOnj1b5voZGRmYNGkSxGIxjI2NYW9vj+jo6Aq/PmOMCaFC4ZqQkICzZ8/C0dERU6ZMgVgsxuTJkxEfH69RWZGRkQgICEBQUBDi4+PRrl07eHl5IT09vcT1ZTIZevfujbt372LPnj1ISkrCunXrYGNjo+luMPbGkUgkuJ714udmdqryuZvZqSrPxcfHqz34xUpW4T5XFxcXuLi4YPny5fj+++8xY8YMrF27Fk5OTvDz84OPjw9EIlGZZaxYsQLjxo2Dj48PACAsLAxRUVHYsGEDZs6cWWz9DRs24MmTJzh9+jQMDZ/fP8jOzq6iu8DYGyU8PBzz4yNKfI5vG6N9FQ7X/Px87Nu3Dxs3bsThw4fRsWNHjB07Fg8ePMCsWbNw5MgRbN++vdTtZTIZzp8/j8DAQOUyPT099OrVC3FxcSVus3//fnh6emLSpEn45Zdf0KBBAwwfPhwzZsyAvr5+idtIpVKVbovMzMwK7jFjus3X1xfWB9RrjfJtYypP43CNj4/Hxo0bsWPHDujp6cHb2xsrV67EW2+9pVzngw8+QIcOHcos59GjR5DL5bC0tFRZbmlpicTExBK3uX37Nv744w+MGDEC0dHRuHnzJiZOnIj8/HwEBQWVuE1wcDDmz1dv+gljrzOxWAz72uoFJt82pvI0DtcOHTqgd+/eWLt2LQYOHKj8ev6ypk2bYujQoVqp4MsUCgUaNmyIiIgI6Ovrw9XVFSkpKVi6dGmp4RoYGIiAgADl48zMTNja2mq9bowx9jKNw/X27dto0qRJmevUrFkTGzduLHMdCwsL6OvrIy0tTWV5WloarKysStxGLBbD0NBQpQvA0dERqampkMlkMDIyKraNsbExjI2Ny6wLY4xpm8azBdLT03HmzJliy8+cOYO//vpL7XKMjIzg6uqKmJgY5TKFQoGYmBh4enqWuE3nzp1x8+ZNKBQK5bLr169DLBaXGKyMMVZVNA7XSZMm4f79+8WWp6SkYNKkSRqVFRAQgHXr1mHz5s24du0aPvvsM+Tk5ChnD3h7e6sMeH322Wd48uQJ/P39cf36dURFRWHhwoUavy5jjAlN426Bq1evltjZ3b59e1y9elWjsoYMGYKHDx9i7ty5SE1NhbOzMw4ePKgc5EpOToae3ov8t7W1xaFDhzBt2jS0bdsWNjY28Pf3x4wZMzTdDcYYE5TG4WpsbIy0tDQ0a9ZMZblEIoGBgeYzuyZPnozJkyeX+FxsbGyxZZ6envjzzz81fh3GGHuVNO4WeOeddxAYGIinT58ql2VkZGDWrFno3bu3VivHGGO6SuOm5rJly/C///0PTZo0Qfv27QEACQkJsLS0xI8//qj1CjLGmC7SOFxtbGxw6dIlbNu2DRcvXoSJiQl8fHwwbNiwEue8MsbYm6hCp7/WrFkT48eP13ZdGGPstVHhawtcvXoVycnJkMlkKsv79+9fyhaMMfbmqNAZWh988AEuX74MkUgEIgIA5RWw5HK5dmvIGGM6SOPZAv7+/mjatCnS09NhamqKv//+G8ePH4ebm1uJU6cYY+xNpHHLNS4uDn/88QcsLCygp6cHPT09dOnSBcHBwfDz88OFCxeEqCdjjOkUjVuucrkctWvXBvD84iv//PMPAKBJkyZISkrSbu0YY0xHadxybdOmDS5evIimTZvCw8MDS5YsgZGRESIiIoqdtcUYY28qjcN19uzZyMnJAQB8/fXXeP/999G1a1eYm5sjMjJS6xVkjDFdpHG4enl5KX9v0aIFEhMT8eTJE9SrV6/ce2YxxtibQqM+1/z8fBgYGODKlSsqy+vXr8/ByhhjL9Go5WpoaIjGjRvzXFbGGIDnV8NT9xbc8fHP7ybyptz4UONuga+++gqzZs3Cjz/+iPr16wtRJ8aYjggPD0dEhHo3AI2IwBt1u26Nw3XNmjW4efMmrK2t0aRJE9SsWVPl+fj4eK1VjjFWvfn6+kIieXHKe35+HjZu7AIA8PE5CUNDk5fWxRvTagUqEK4DBw4UoBqMMV1U9Gu+TJaj/N3KyhlGRi8aX2/a3bo1DtfSbmHNGHszdYudp/w9Ty5D8H+/dz3xLUz0X75x6Dy8SSp8VSzGGJNIJLie9WJAS6rIV/5+MzsVxnovrvFsFh/PA1pl0dPTK3PaFc8kYOzNER4ejvnxESU+55ewUXWBawQPaJVl3759Ko/z8/Nx4cIFbN68GfPnqzdqyBh7Pfj6+sL6gHpTsdzW+b4xrVagAuE6YMCAYssGDRqE1q1bIzIyEmPHjtVKxRhj1Z9YLIZ9bfUC0+UNG9HS+KpYpenYsSNiYmK0VRxjjOk0rYRrXl4evvvuO9jY2GijOMYY03kadwsUvUALESErKwumpqbYunWrVivHGGO6SuNwXblypUq46unpoUGDBvDw8EC9evW0WjnGGNNVGofr6NGjBagGY4y9XjTuc924cSN2795dbPnu3buxefNmrVSKMcZ0ncbhGhwcDAsLi2LLGzZsiIULF2qlUowxpus0Dtfk5GQ0bdq02PImTZogOTlZK5VijDFdp3G4NmzYEJcuXSq2/OLFizA3N9dKpRhjTNdpHK7Dhg2Dn58fjh49CrlcDrlcjj/++AP+/v4YOnSoEHVkjDGdo/FsgQULFuDu3bvo2bMnDAyeb65QKODt7c19rowx9h+Nw9XIyAiRkZH45ptvkJCQABMTEzg5OaFJkyZC1I8xxnRSha/n2rJlS7Rs2VKbdWGMsdeGxn2uH330ERYvXlxs+ZIlSzB48GCtVIoxxnSdxuF6/PhxvPfee8WW9+nTB8ePH69QJUJDQ2FnZ4caNWrAw8MDZ8+eVWu7nTt3QiQS8X29GHtNPL9Vd7wGP+pdS7YqaNwtkJ2dDSMjo2LLDQ0NkZmZqXEFIiMjERAQgLCwMHh4eCAkJAReXl5ISkpCw4YNS93u7t27mD59Orp27arxazLGqidNbtUNAGJx9b2zgcbh6uTkhMjISMydO1dl+c6dO9GqVSuNK7BixQqMGzcOPj4+AICwsDBERUVhw4YNmDlzZonbyOVyjBgxAvPnz8eJEyeQkZGh8esyxqofTW7V/Xz96ntnA43Ddc6cOfjwww9x69Yt9OjRAwAQExOD7du3Y8+ePRqVJZPJcP78eQQGBiqX6enpoVevXoiLiyt1u6+//hoNGzbE2LFjceLEiTJfQyqVQiqVKh9XpHXNGHs1xGIxhiWFKx/nyWUovBPX4BtRRe4mC4jF815d5TSkcbj269cPP//8MxYuXIg9e/bAxMQE7dq1wx9//IH69etrVNajR48gl8thaWmpstzS0hKJiYklbnPy5EmsX78eCQkJar1GcHAw39uLMR2hyd1kAcBBIqm29+Wq0FSsvn37om/fvgCetwR37NiB6dOn4/z584Le/TUrKwsjR47EunXrSrx4TEkCAwMREBCgfJyZmQlbW1uhqsgYqwSN7iYL4J9w8evT51ro+PHjWL9+Pfbu3Qtra2t8+OGHCA0N1agMCwsL6OvrIy0tTWV5WloarKysiq1/69Yt3L17F/369VMuUygUAAADAwMkJSWhefPmKtsYGxvD2NhYo3oxxqqGJneTBYB+vr4C1qZyNArX1NRUbNq0CevXr0dmZiY+/vhjSKVS/PzzzxUazDIyMoKrqytiYmKU06kUCgViYmIwefLkYuu/9dZbuHz5ssqy2bNnIysrC6tWreIWKWM6TpO7yRauX12pHa79+vXD8ePH0bdvX4SEhODdd9+Fvr4+wsLCKlWBgIAAjBo1Cm5ubnB3d0dISAhycnKUswe8vb1hY2OD4OBg1KhRA23atFHZvm7dugBQbDljjFUltcP1t99+g5+fHz777DOtnvY6ZMgQPHz4EHPnzkVqaiqcnZ1x8OBB5SBXcnIy9PS0dgdwxhh7JdQO18JReldXVzg6OmLkyJFau8Tg5MmTS+wGAIDY2Ngyt920aZNW6sAYY9qkdpOwY8eOWLduHSQSCXx9fbFz505YW1tDoVDg8OHDyMrKErKejDGmUzT+vl2zZk2MGTMGJ0+exOXLl/H5559j0aJFaNiwIfr3719+AYwx9gaoVGemg4MDlixZggcPHmDHjh3aqhNjjOk8rYwU6evrY+DAgdi/f782imOMMZ3Hw/CMMSYADlfGGBMAhytjjAmAw5UxxgTA4coYYwLgcGWMMQFwuDLGmAA4XBljTAAcrowxJgAOV8YYEwCHK2OMCYDDlTHGBMDhyhhjAuBwZYwxAXC4MsaYADhcGWNMAByujDEmAA5XxhgTAIcrY4wJgMOVMcYEwOHKGGMC4HBljDEBcLgyxpgAOFwZY0wABlVdAcYYE5JEIoFEIlFr3fh4QCwWQywWV/p1OVwZY6+18PBwRETMV2vdiAggKCgI8+bNq/Trcrgyxl5rvr6+sD7wouUqVeTDL2EjAOA7Zx8Y6xkqn3Nb56uVVivAfa6MMSYIbrkyxl5r4eHhmB8fUeJzhS1YJdcI7hZgjDF1FO0WKIs2uwU4XBljrzWxWAz72uoFpouLi9Zel/tcGWNMANUiXENDQ2FnZ4caNWrAw8MDZ8+eLXXddevWoWvXrqhXrx7q1auHXr16lbk+Y4xVhSoP18jISAQEBCAoKAjx8fFo164dvLy8kJ6eXuL6sbGxGDZsGI4ePYq4uDjY2trinXfeQUpKyiuuOWOMla7Kw3XFihUYN24cfHx80KpVK4SFhcHU1BQbNmwocf1t27Zh4sSJcHZ2xltvvYUffvgBCoUCMTExr7jmjDFWuioNV5lMhvPnz6NXr17KZXp6eujVqxfi4uLUKiM3Nxf5+fmoX79+ic9LpVJkZmaq/DDGmNCqNFwfPXoEuVwOS0tLleWWlpZITU1Vq4wZM2bA2tpaJaBfFhwcjDp16ih/bG1tK11vxhgrT5V3C1TGokWLsHPnTuzbtw81atQocZ3AwEA8ffpU+XP//v1XXEvG2JuoSue5WlhYQF9fH2lpaSrL09LSYGVlVea2y5Ytw6JFi3DkyBG0bdu21PWMjY1hbGyslfoyxpi6qrTlamRkBFdXV5XBqMLBKU9Pz1K3W7JkCRYsWICDBw/Czc3tVVSVMcY0UuVnaAUEBGDUqFFwc3ODu7s7QkJCkJOTAx8fHwCAt7c3bGxsEBwcDABYvHgx5s6di+3bt8POzk7ZN1urVi3UqlWryvaDMcZeVuXhOmTIEDx8+BBz585FamoqnJ2dcfDgQeUgV3JyMvT0XjSw165dC5lMhkGDBqmUo62LLTDGmDZUebgCwOTJkzF58uQSn4uNjVV5fPfuXeErxBhjlaTTswUYY6y64nBljDEBcLgyxpgAOFwZY0wAHK6MMSYADlfGGBMAhytjjAmAw5UxxgTA4coYYwLgcGWMMQFwuDLGmAA4XBljTAAcrowxJgAOV8YYEwCHK2OMCYDDlTHGBMDhyhhjAuBwZYwxAXC4MsaYADhcGWNMAByujDEmAA5XxhgTAIcrY4wJgMOVMcYEwOHKGGMC4HBljDEBcLgyxpgAOFwZY0wAHK6MMSYADlfGGBMAhytjjAmAw5UxxgTA4coYYwLgcGWMMQFwuDLGmAA4XBljTAAcrowxJoBqEa6hoaGws7NDjRo14OHhgbNnz5a5/u7du/HWW2+hRo0acHJyQnR09CuqKWOMqafKwzUyMhIBAQEICgpCfHw82rVrBy8vL6Snp5e4/unTpzFs2DCMHTsWFy5cwMCBAzFw4EBcuXLlFdecMcZKZ1DVFVixYgXGjRsHHx8fAEBYWBiioqKwYcMGzJw5s9j6q1atwrvvvosvvvgCALBgwQIcPnwYa9asQVhYWLH1pVIppFKp8vHTp08BAJmZmSrr5RRIoa6i25ZG3TLVLU+IMnm/tVOeEGXysXn1ZRYtr/AxEaldJyWqQlKplPT19Wnfvn0qy729val///4lbmNra0srV65UWTZ37lxq27ZtiesHBQURAP7hH/7hnwr/3L9/X+N8q9KW66NHjyCXy2Fpaamy3NLSEomJiSVuk5qaWuL6qampJa4fGBiIgIAA5WOFQoEnT57A3NwcIpGozPplZmbC1tYW9+/fh5mZmTq79ErLE6JMXaijEGXqQh2FKJPrWHaZRISsrCxYW1tr/BpV3i0gNGNjYxgbG6ssq1u3rkZlmJmZae2gClGeEGXqQh2FKFMX6ihEmVzH0tWpU6dCZVfpgJaFhQX09fWRlpamsjwtLQ1WVlYlbmNlZaXR+owxVhWqNFyNjIzg6uqKmJgY5TKFQoGYmBh4enqWuI2np6fK+gBw+PDhUtdnjLGqUOXdAgEBARg1ahTc3Nzg7u6OkJAQ5OTkKGcPeHt7w8bGBsHBwQAAf39/vP3221i+fDn69u2LnTt34q+//kJERITW62ZsbIygoKBi3QrVpTwhytSFOgpRpi7UUYgyuY7aK7MYjYfABLB69Wpq3LgxGRkZkbu7O/3555/K595++20aNWqUyvq7du0ie3t7MjIyotatW1NUVNQrrjFjjJVNRFSRCVyMMcbKUuVnaDHG2OuIw5UxxgTA4coYYwLgcK0AbXVTFxQUaKWcQrm5uVotT4gys7OztVpeIT4mry9dOCYl4XBV05MnTxAdHY1r166pXAimIogIZ86cwbp16/Do0aNK1y0/Px++vr6YOHEiJBJJpcsDntfx6NGj+OKLL5CSklLp8nJzc7FlyxZ89NFHSE5O1kIN+ZhUVmX/z4SmC8ekLByuakhISEDfvn0xZcoUfPTRR5gxY0alyhOJREhPT8cvv/yC6dOnV7p+hoaGGDhwIPLy8uDr61vp8oDndbS1tcWTJ08qVaZCocDx48fx6aefYuXKlTh+/DiGDx9e6frxMal4mc+ePcOcOXPQv39/ODk5wcfHB7/99ptW6hgTE4ORI0di8eLF+PPPPytVli4ckzJV2SSwak6hUNDFixcpOTmZVq9eTR06dKCTJ0/S6dOnycLCgg4cOFChcqVSqfL3yMhIEolEdOzYsQrVr6jbt29Tw4YN6ffff69Q3QrJZDLl7/fv36cGDRpQdHR0hcq6c+cOTZs2jQwNDengwYMklUqpTp069PPPP2tcFh+T5yp6TBQKBW3dupX09PSoZs2aNHv2bNqxYwcFBARQs2bNKnyMiYgKCgpo0qRJJBKJqG/fvuTu7k7t2rWr8H7rwjEpD4drCRITE8nGxoZcXFyocePG5OLiQmFhYcrnv/nmG+rSpYtGZebn56s8XrBgAZmamlLbtm3p4MGDapeTl5dXbJlcLleWP2LECJoyZYpGdXvy5IlKPWfNmkVHjx5VLvP396dBgwZVqDwiooSEBGrVqhXt2rWLiJ7//1lbW2tURz4mlTsm2dnZNGjQIGrQoAE5OztT165dqXfv3nTq1Ckiev7/16xZM43q+LJ79+5Rs2bNlMckKyuLli9frnGZ1fmYaIrDtYiCggLq2rUrffLJJ3Tv3j2Kj4+ncePGkb29vXKdb7/9lnr06EGPHz8us6w7d+7QzZs3VZbt3r2brK2tqUGDBhQcHEx37twp9gdVWr169epFLVq0oEmTJtHmzZvp1q1bxdZzcnIqdr3b0igUClq5ciVNmzZNpWU0bdo0at68ufLxRx99RJMmTSqxFVC0vIULF9LkyZNVyiMiWrNmDXl4eCj31dnZWe2WEh+Tih+TQitXriRbW1v64YcfKDc3l3Jzc2n16tVkY2NDMpmMnj59Sq1atdLobMeHDx9Sbm4uERH98ssvZGNjQ0+ePCGFQkEKhYL++ecfcnR0pNOnT5dZTnU/JhXF4VpEYmIitWnThnbu3KlclpqaSrVq1aKxY8fS+PHjydnZmaZNm1ZuWUFBQeTk5ERERFeuXCFPT08yMDCg8ePH04ULF1S++qjjs88+I5FIRF9++SU1b96cLC0tqXPnzjR8+HCaN28eubu7U506dejMmTNqlzlv3jzq3bs3ET1/Y8vlcrp+/TqZmprShx9+SE5OTlSzZk3avn27WuV99dVX1KNHD2V5BQUFRES0atUq6tSpk7JF9vDhQ7XryMek5GOybds2tcrLz8+nAQMG0OTJk1WWp6SkUMeOHenUqVOUn59PR44coeTkZLXKlEgktHTpUkpISCAiomfPnpGBgYHKV/eDBw9S06ZNKTU1tcyydOGYVASHaxEymYxatmxJkZGRymUSiYRcXV1p+fLl9OGHH9KECROUf1Rlyc3NJVNTU3JzcyN9fX3q3bs3/f7775SZmVls3YKCgmKf3kXl5eWRsbGxsq/or7/+ok2bNtG4cePI3d2dxowZQ2fPntVof/Py8sjMzIz27NmjXHbhwgVq06YNHTp0iPz8/CgyMlLtP/DC8nbv3k1yuVy5vHfv3tSxY0eVcp49e6b8WlqW8o7JwIEDacKECfTXX38RESkDvSSaHBN1vKpjkpCQQE5OTsWOyb179yg0NLTM8goKCsjV1ZXWrFmjsjwsLIzq1q1LN27cICJStkLVkZ+fTw4ODrRu3Trlsq+//pratGlDu3btojVr1pC7uzt5e3sTEan8LRSlC++TiuBwLcHy5cvJ3t6eEhIS6NatW7RhwwZycHAgiURCCoVCra8nhX744QcSiUS0atUqSktLU3mu8Ctdfn4+bdu2jQYMGEDXr18vs7xFixaRnZ2dWq998+ZNysrKKne91atXU5s2bWjdunX066+/0pAhQ6h3797KfquX3xjqfA1du3YttWrVinx9fSkkJIS6detGDg4OKkG6f/9+6tOnD9WqVYuuXLlSbpllHZOX+9LU8fIxSU1NLbZPha1FdfdXk2Ny+vTpYn8HJXn5mERFRdHQoUPpf//7Hz179kxlvRs3bpBIJKL4+Pgyy1u+fDm1bNmSzpw5QwkJCbR9+3bq1q0bjRs3rtgHZ0pKilpBu2zZMmrRooXK36y/vz81b96cRCIR9ejRQ+2WcFW+T9Q9JpricC3FkCFDqFmzZtS6dWuysrIiPz8/kslkavdxvaxJkybk5+dX5jr379+nTz/9lLp27VpueQ0bNqR58+YRESn7t4hUQ/DcuXM0fvx4mjt3rlp1/Pbbb+mtt94iS0tL6tatG61YsYJu375NiYmJRETKkFZ3/zdv3kxdunShRo0a0bBhwygyMpJkMhndv3+fpkyZQg4ODjR27Fhyd3en9u3bq1VmScekMBiePXtGwcHB5OzsTGPGjKG9e/eWWVbRY/L48WPKyMigf/75R626FFXaMXk59K9cuUIjRoyg4cOHq1Xmy8ekc+fOtHv3bpXnC1vo/v7+5OLiUm5548ePp5YtW1Lz5s3Jzs6O3n//fbp9+7bKOn///TcFBgbSwoUL1arje++9R23atKFBgwbRgAEDqEePHtSoUSMaO3YspaWlUWJiIo0dO5bmzJlT7oDUq3qfVOaYaILDtRRPnjyhkydPUkhIiEajlCU5ePAgubi4UHp6uspyhUJBKSkpysePHj0ia2tr+umnn8os7/fff6cJEyYUGzQqLLPw3/DwcGrcuHG5QVMoJSWFLl26RPHx8TR37lyytramnj17UseOHaljx47UtWtXmjVrFs2ePZtOnDhRbnm5ublUUFCgMsiUmJhIEydOJGtra3rw4AEREZmbm9OPP/5YbnmFx2TlypUqxyQnJ4d69epFderUoS+//JKGDRtGDg4OKl+rizp48CB16NCBUlNTSSaT0bJly8ja2preeecd6ty5Mw0cOJD69u1Lq1evplWrVpXbDXTo0CHy9fUtswWtUCjoypUrZG1tTTt27Ch3f4meH5OLFy/S1atXi5VVKDc3l+rVq1duH2xeXh4lJibSkSNH6OTJkyplFZYnk8lozZo11Lx583L/DomI0tLSaP/+/TRgwAAaPnw4ffXVV3Tu3Dnl8+7u7uTl5UWenp701ltvlVmmtt8nQh0TdXG4viJXr16l7Oxsla9vubm5NHLkSJX5mePHj6fRo0eXW55EIin1ucI3S1RUFBkZGZGDg4PG9f3xxx9JJBJRdnY2XbhwgRITE+nkyZO0bds2EolEKiP1ZUlMTFS+2QrfwA8ePKDWrVtTeHg4ERGtWLGCLCwsNK4j0fPW+oULF8jGxoZiY2OVy9etW0e2trZlbvvysdi6dSuJRCKSSCR0+PBhiomJod27d9OiRYtIJBKpjNSXpnDgpuiHXmJiosqI9dq1a8nR0VGt/VPXggULSCwWa7zdy8FTeHwuXbpEFhYWah3jot9kMjIy6JtvvqGFCxfSrl27qEuXLvTnn3/Ss2fPaO3atdSoUaMyy9P2+6QqjwmH6yuSn59P48ePp4CAAJUBl9mzZ5ONjY2ydefh4aH2V3mi5/MXXyaXy+nKlSs0ZswY0tfXp/fee69Yq0cdubm51L59e/L19VUu27hxI5mbm5ObmxutWrWq3P7cvLw8GjBgAM2ePVuldUREtGTJEnrvvfeUf/QBAQF06dIltet348YN+vfff4mIaOfOncp5s1KplORyOSUmJlLbtm3p4sWLau+vi4uLyv7++uuv1L59e+rWrRv5+/uXO+pN9Lz75KefflKZDbF3715q166d8vHJkyfJycmJLl++rFbdXnbz5k1KSUmhn3/+mUJDQ2n69OnUpUsX6t69O4lEItq6data5RTtuyUiyszMpDlz5pCJiQl5enrSgQMHyhwcLOrUqVNkaWlJLVq0oGHDhpGzszPZ2Ngon8/KyiInJyf65ZdfSi1DiPeJ0MekNByur9D8+fOVfUVyuZwKCgro0qVLVKdOHXr33Xepdu3aVLduXbVG0Imej4J+8cUXysGHlJQU+vbbb6levXrUvHlz2r9/v3JdTfqKC9eNi4sjAwMD2rp1K7Vu3ZpMTU1p+vTpdO3aNbVnD3z11VfUuXNn5ePCfuFRo0ZR//79lY81mW5z48YNmjJlijKMZTIZGRsbq8zR3LBhA9nb2xf78ClJ4f6eOnWKDAwMaO/evdS3b19q1KgRjR49mn777bdiJ0aUJioqitzc3FQGXLKyssjKyoqmTZtGN27coLlz51KzZs3o6dOnapV55swZGjx4MLVv357atWtHZmZmZGZmRq1atSJHR0f67LPPKCAggPbs2aPWh8mtW7do2bJlKh+OP/74I9na2pKlpSWtWbNGrYHQogoHzQoHCVNSUqhJkyY0f/58IiKKjY2lBg0a0B9//FFmOdp+nwhxTNTB4foKKRQKMjc3p82bNyvf0DExMeTk5ETXrl2j8PBwtUbOX1ajRg1asGAB7d69m1q1akU1a9akxYsXF3vdipBIJNSoUSMSiUQ0YsQIOnHiBOXk5GhUhkKhIAsLC1q7dq2y5ZCUlER2dnZqzUstzctnAxERhYaGkouLCwUFBdHEiROpQYMGyje1uh49ekQtW7akGjVq0MCBA2nLli3KfmFN2NnZ0ZYtW4joeUtMLpfTxx9/TJ07d6YWLVqQkZERhYSEEJH6x6Z37940ZcoUWrlyJe3Zs4dSUlJKHDEv2j9ZGiMjI1q5ciWdO3eOunTpQkZGRuTn50f379/XYE9VTZgwgT755BOVZUuXLqWGDRuSh4cHtWzZkt59991yP6jKe5+EhYVp3MLU5Jho0lovC4frK7Z582Zq1aoVffHFF7RixQpycnKiYcOGVbi8iIgIEolEZGBgQKNHj1YZPCprbmFpCv+YJ02aREZGRtSmTRuqUaOGygR+Te3cuZNcXV2pXbt29P7775O+vj55eXlVqHVUaO3atdS4cWOV1svUqVOV92KbOHGiWtNrCvf366+/ppYtW1Lbtm3J2Ni42Nk7mvxfrl27lpo2bUpJSUnKZW3btqVdu3bR9evX6dGjRxUuuyQvD2Lm5OTQ+++/rzJgVZKwsDASiURkaGhI7733Hp0/f77YOpqGzNatW6lp06Yqrb9PPvmEJk6cSPv376fNmzer3B+vpAHZQiW9T4YOHapRfV6m6THRBr6HVhXYsmULIiIi8M8//8DT0xMBAQFwdXWtcHmenp6wtrbG3r17AQByuRz6+vqVqmP//v3RsWNH+Pj4YNeuXdDT08OUKVMqXN65c+ewY8cO3LlzBx988AH69u0Lc3NzKBQK6OlV7OJsY8aMwcmTJ1G3bl08fvwYZmZmqFmzJj788EOMHTsWderUUbusIUOGwNzcHD4+Pti/fz9q1apV7EpbRASRSKTW/+9HH32E1NRUiMViXL58GSKRCEePHoVYLFauU1BQgNjYWJw+fRpz584tt46Fr1tYj9JMmzYNv/76K27cuFFmeZ06dYKJiQkOHToEA4PSbwT99OlTtf8vXVxc0L59e3h6ekImk2HDhg348ssv8fHHH6usFxcXhxkzZuD48eOlllX0ffLFF1/A2dkZGRkZSEtLw5EjR5CWlgYHBwf07NkTVlZWZdZNnWNS+LpRUVGIjIxUa59LpfW4ZmrJysqizMxMrXxiHj16VK2J5JrIy8srs2VRHWRnZ1NCQgLNmTOHgoOD6ccff6S///67QmVlZGSotPplMhmFhYXR9u3bacuWLXT+/Hm6e/cuJScnq5zWW5q0tDTatm0beXp60qRJkygmJoaInrdSX26p7tixg1q0aKH8yqoNeXl5ZGlpqdJtUpJjx46RSCQqNs1MKpXS1atXac2aNeTm5kZdunShgIAAtU4X/fvvv2n48OFkbm5O7u7uNGLEiGJTq4ieD6iZmppSREREmeVlZWXR06dPlYOJSUlJNHv2bGrZsiWZm5vTmDFjqFmzZtSzZ88yp94RlX5MCl24cIFcXFzI1NSURCIRrV+/vtz9LQuH62ti0aJFdPz4cUFfo6J9t0W3r2w5RcsTQmxsLJmamlK3bt3I19eX7O3tycvLi1q3bk3Dhg2jTp06Var8wilQCoWC1q9fX6mpaETFv8Lv3buXzMzMyt1+5syZtHv3bmV9MjIyaO3atWRvb0+1atWizz77jGbMmEHvv/8+NW7cWK065eXl0Z07d0rtFy08bps3b6batWurVSbR8+lTrq6upK+vTyKRiAICApTdH4sXL6YGDRqoXVZRJ06cICsrK7KysqI7d+7Q999/r9b/X1k4XF8Tcrlc0LCp7rS977m5udSzZ08aNWqU8jHR8ys4ubm5kUgkUvvkjDt37iinjSkUCkpOTi52JpKHh0e5LU113L17lx4/fky//vor1ahRgz7//PMy1y8c3Cnk7e1NBgYG5OvrW2zw0sXFhRYtWqRxnQq/AZV0GrWrqyv5+/uXW0ZSUhI1a9aMOnbsSNHR0XTgwAFavHgxicVi5TeODh060IIFC9Sq0+3bt1UG1h4/fkybNm0iOzs7ZUu7ffv29Omnn6pVXkk4XBkrxYkTJ8jAwEA55WvJkiVkZWVF9vb2FBkZqdY0rydPntDixYvpt99+Uy6TSqXk6upKs2bNIqLngeji4lLuV+SXnT9/nq5evUr79u2j2bNnk5+fH7Vt25bat29PRkZGZGtrS9bW1tSjRw+1unfy8/NpypQpJBaLVb4BPXv2jBQKBeXl5VH37t3p22+/VbuORCUP1hVekOXWrVu0fPlyEolE5c5w2LVrF9na2qp0+xQUFFDPnj2Vs0K+++47Wrx4cbkftA8fPqT58+cXu7yiTCYjLy8v5SyWEydOUOvWrSs8Pav0XmzG3nBdunTBqFGjMHjwYJiYmODmzZsICAjAlClTYGFhoVYZ9erVQ3R0NPLy8vDuu+9CoVDAyMgI/v7+8PX1xaNHj2BoaIj09HQ4OjqqXbfPP/8c8fHxyM/Ph5OTE+RyOXr16oX69etj3rx5sLe3h4GBAVq0aKFWeTKZDCdPnsSkSZPQtWtX5Ofng4hgbGwM4Pltdc6fPw9/f3+165idnY3MzEzk5uYiLi4OUqkUR44cgUwmw6VLl5CZmQkzMzMAwMGDBzFmzJhSy4qPj4e7uztatWqlHATV19eHVCpVDmR5e3urNfBmYWGB6OhoyOVyvPfee8qBQkNDQzRr1gx5eXmQSqXo0qUL9u7dW+ZgX5kqFMmMvQGSkpKod+/eJBKJyNPTk/79998KdT/s2LGDrKysVAbM1q5dSx4eHrRixQrq1q0bfffddxqV+ffff9O+ffsoKyur3FafOnW+du0aNW/evNhZclKplBYvXkz169dXXj5QXfHx8dSwYUMyMDAgR0dHsrCwoHfeeYdGjBhB8+bNo+joaDp8+DDl5+eXe5Hz3bt3k1gsVs7DlcvltHDhQhKJRHTo0CGVddWZ2rZ9+3YSi8UqZ92lpKSQg4MDffXVVxrtZ2l4KhZjJYiIiMCECRPQtWtXtGzZEmfOnMHly5crXF6XLl3g6OiIIUOGwNXVFWPHjoW5uTnWrVuH/Px8GBoaAkC506zKou5UrdK4urqiQ4cO8PPzQ6tWrbBt2zZs2LABT548gZeXF6ZOnQorKyuNyl+/fj3atWsHkUgEe3t71K5du8JTBfv06YOsrCyYm5vjwYMHyM/Px8iRI/HFF18AgLJFq+70vu7du8PCwgKdOnWCg4MDfvjhB/z555/Yu3cvPD09VdbNyclBzZo1NauwViKasddMUlISff/995STk0PPnj2jDRs2UGZmZoUHzm7cuEGDBw8mKysrqlWrFllZWSkv7k1U+RMJtOHSpUvk7OxMTZo0obp165KlpSUNHz6cNm3apLVJ9oX7WZH9TU9Ppz179lC/fv1o1qxZytO7yzp1uqzjdffuXQoICCArKysyNzenNm3aUFxcXLH1Tp06RePGjVPpN1cHt1wZe0Vyc3Nx8uRJZGZmok+fPqhZs2alWqpCSEtLw/nz55GRkYF27dqhfv36xSbZa0rIfVywYAESEhJgbGyM5s2bQ6FQwNPTE1KpFO7u7rCxsSm3FXvv3j38+++/cHZ2BvC8BVxYX5FIhKSkJISGhiImJgYJCQnKbxnl4XBljOkchUIBhUIBb29vxMTEIDQ0FPfu3UNycjIePHgAuVyOI0eOYOjQofjhhx/ULrewy6Jo18KiRYswa9YsLFmyBNOnT1erLJ4twBjTOXp6etDT08P06dOxZ88etG3bFoMGDQIAJCUl4fvvv0dubi4yMjJQUFCg9oi/SCQCESmD9fDhwwgLC0NycjKGDx+O5s2bq11HbrkyxnTa5MmTERsbi3PnziEkJARLly5FrVq1EBoain79+qlVxuXLl3H69Gn4+voCeB7Q3333HeLi4tCoUSP06dMHAwYMgLW1tdr14nBljOm0pKQkuLu7w8zMDP/++y9mzpyJ2bNnK59Xp893165dGDp0KM6ePYuYmBjs27cPNWrUQPfu3fHBBx+gbdu2GpUHcLgyxnRUQUEBTp06hZUrV2L//v0AgPT0dOUJHppecc3LywtHjx6Fm5sb3NzcMGDAAPTo0UMZpJoOzFXsWm+MMVbFnj59iu7duyMlJQWxsbHo06cPjh49qnxe00tZfv/99ygoKMD//vc/BAcHo2fPnhCJRFAoFACg8YwHbrkyxnTWhQsX0L59ewDA/fv3YWBgUKmpY76+vvjrr79w7tw5iESiSk0h43BljOk8bc2llclk6Ny5M7Zv346WLVtWqiwOV8YYe4k27uQBcLgyxpggeECLMcYEwOHKGGMC4HBljDEBcLgyxpgAOFwZY0wAHK6MMSYADlfGGBMAhyurtrp164apU6dWdTWUiAjjx49H/fr1IRKJkJCQUNVVgp2dHUJCQpSPU1NT0bt3b9SsWRN169atsnoxvlg2Y2o7ePAgNm3ahNjYWDRr1kzt22trw6ZNmzB16lRkZGSoLD937pzKjfNWrlwJiUSChIQEtW4zzYTD4creKHK5HCKRSOMrJgHArVu3IBaL0alTJwFqVjENGjRQeXzr1i24urpW+rx4VnncLcDK1K1bN/j5+eHLL79E/fr1YWVlhXnz5imfv3v3brGvyBkZGRCJRIiNjQUAxMbGQiQS4dChQ2jfvj1MTEzQo0cPpKen47fffoOjoyPMzMwwfPhw5Obmqrx+QUEBJk+ejDp16sDCwgJz5szBy2dsS6VSTJ8+HTY2NqhZsyY8PDyUrws8b/HVrVsX+/fvR6tWrWBsbIzk5OQS9/XYsWNwd3eHsbExxGIxZs6ciYKCAgDA6NGjMWXKFCQnJ0MkEsHOzq7EMgpf79dff4WDgwNMTU0xaNAg5ObmYvPmzbCzs0O9evXg5+cHuVyu3O7ff/+Ft7c36tWrB1NTU/Tp0wc3btxQ/v/5+Pjg6dOnyis1FR6Dl7sF7OzssHfvXmzZsgUikQijR48GEWHevHlo3LgxjI2NYW1tDT8/v9ION9Mmje4Vy944b7/9NpmZmdG8efPo+vXrtHnzZhKJRPT7778TEdGdO3cIAF24cEG5zb///ksA6OjRo0REdPToUQJAHTt2pJMnT1J8fDy1aNGC3n77bXrnnXcoPj6ejh8/Tubm5rRo0SKV165Vqxb5+/tTYmIibd26lUxNTSkiIkK5zqeffkqdOnWi48eP082bN2np0qVkbGxM169fJyKijRs3kqGhIXXq1IlOnTpFiYmJlJOTU2w/Hzx4QKampjRx4kS6du0a7du3jywsLCgoKIiIiDIyMujrr7+mRo0akUQiofT09BL/vwpfr3fv3hQfH0/Hjh0jc3Nzeuedd+jjjz+mv//+mw4cOEBGRka0c+dO5Xb9+/cnR0dHOn78OCUkJJCXlxe1aNGCZDIZSaVSCgkJITMzM5JIJCSRSCgrK4uIiJo0aUIrV64koue3nn733Xfp448/JolEQhkZGbR7924yMzOj6OhounfvHp05c0bl/48Jh8OVlentt9+mLl26qCzr0KEDzZgxg4g0C9cjR44o1wkODiYAdOvWLeUyX19f8vLyUnltR0dHlXvPz5gxgxwdHYmI6N69e6Svr08pKSkq9evZsycFBgYS0fOwA0AJCQll7uesWbPIwcFB5bVCQ0OpVq1aJJfLiYho5cqV1KRJkzLLKXy9mzdvquyXqampMhCJiLy8vMjX15eIiK5fv04A6NSpU8rnHz16RCYmJrRr1y5luXXq1Cn2ei+HKxHRgAEDaNSoUcrHy5cvJ3t7e5LJZGXWm2kfdwuwcr18/yAAEIvFSE9Pr1Q5lpaWMDU1RbNmzVSWFS23Y8eOKtfp9PT0xI0bNyCXy3H58mXI5XLY29ujVq1ayp9jx47h1q1bym2MjIyK7UNR165dg6enp8prde7cGdnZ2Xjw4IFG+2lqaqpyl1BLS0vY2dmhVq1aJe7rtWvXYGBgAA8PD+Xz5ubmcHBwwLVr1zR67aIGDx6MvLw8NGvWDOPGjcO+ffuUXR1MWDygxcplaGio8vjlW18UDgzRS/2g+fn55ZYjEonKLFcd2dnZ0NfXx/nz54tdf/PlIDMxMdHKhZTVVdJ+VXZfK8rW1hZJSUk4cuQIDh8+jIkTJ2Lp0qU4duxYsTox7eKWK6uUwtFqiUSiXKbN+Z9nzpxRefznn3+iZcuW0NfXR/v27SGXy5Geno4WLVqo/FhZWWn0Oo6OjoiLi1P5kDh16hRq166NRo0aaWVfynrtgoIClX19/PgxkpKS0KpVKwDPW98vD4BpwsTEBP369cN3332H2NhYxMXF4fLly1qpOysdhyurFBMTE3Ts2BGLFi3CtWvXcOzYMZXbGldWcnIyAgICkJSUhB07dmD16tXw9/cHANjb22PEiBHw9vbGTz/9hDt37uDs2bMIDg5GVFSURq8zceJE3L9/H1OmTEFiYiJ++eUXBAUFISAgoELTtjTRsmVLDBgwAOPGjcPJkydx8eJFfPLJJ7CxscGAAQMAPJ8JkJ2djZiYGDx69KjYrIrSbNq0CevXr8eVK1dw+/ZtbN26FSYmJmjSpImQu8TA4cq0YMOGDSgoKICrqyumTp2Kb775Rmtle3t7Iy8vD+7u7pg0aRL8/f0xfvx45fMbN26Et7c3Pv/8czg4OGDgwIE4d+4cGjdurNHr2NjYIDo6GmfPnkW7du0wYcIEjB07VqsfFGXZuHEjXF1d8f7778PT0xNEhOjoaOVX906dOmHChAkYMmQIGjRogCVLlqhVbt26dbFu3Tp07twZbdu2xZEjR3DgwAGYm5sLuTsMfJsXxhgTBLdcGWNMAByujDEmAA5XxhgTAIcrY4wJgMOVMcYEwOHKGGMC4HBljDEBcLgyxpgAOFwZY0wAHK6MMSYADlfGGBPA/wECnk2FSsLwSgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 325x325 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading filename='data_cache/2023-09-27_scan_N_SM_time.json'...\n",
      "     N_SM_time  accuracy        TP        TN  method      time\n",
      "0            1  1.000000       NaN  1.000000  HD-SNN  0.069434\n",
      "1            1  1.000000       NaN  1.000000  HD-SNN  0.063933\n",
      "2            1  1.000000       NaN  1.000000  HD-SNN  0.059802\n",
      "3            1  1.000000       NaN  1.000000  HD-SNN  0.060135\n",
      "4            1  1.000000       NaN  1.000000  HD-SNN  0.054192\n",
      "..         ...       ...       ...       ...     ...       ...\n",
      "135        127  0.999940  0.966430  0.999970    corr  0.859307\n",
      "136        127  0.999954  0.973301  0.999977    corr  0.850763\n",
      "137        127  0.999951  0.972048  0.999976    corr  0.839471\n",
      "138        127  0.999953  0.973746  0.999977    corr  0.843567\n",
      "139        127  0.999951  0.972443  0.999975    corr  0.892726\n",
      "\n",
      "[140 rows x 6 columns]\n",
      "Bbox(x0=0.125, y0=0.25, x1=0.95, y1=0.975)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x325 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading filename='data_cache/2023-09-27_scan_N_pre.json'...\n",
      "     N_pre  accuracy        TP        TN  method      time\n",
      "0        8  0.998348  0.133626  0.999173  HD-SNN  0.150681\n",
      "1        8  0.998264  0.128367  0.999131  HD-SNN  0.154906\n",
      "2        8  0.998317  0.116146  0.999158  HD-SNN  0.142787\n",
      "3        8  0.998338  0.127912  0.999169  HD-SNN  0.137957\n",
      "4        8  0.998292  0.146923  0.999145  HD-SNN  0.156435\n",
      "..     ...       ...       ...       ...     ...       ...\n",
      "255   2351  0.999998  0.999143  0.999999    corr  2.499372\n",
      "256   2351  0.999999  0.999553  1.000000    corr  2.534012\n",
      "257   2351  1.000000  1.000000  1.000000    corr  2.531489\n",
      "258   2351  0.999997  0.998680  0.999999    corr  2.543407\n",
      "259   2351  0.999998  0.999138  0.999999    corr  2.536328\n",
      "\n",
      "[260 rows x 6 columns]\n",
      "Bbox(x0=0.125, y0=0.25, x1=0.95, y1=0.975)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x325 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading filename='data_cache/2023-09-27_scan_E_SM.json'...\n",
      "     E_SM  accuracy        TP        TN  method      time\n",
      "0       1  0.998129  0.018901  0.999064  HD-SNN  0.334534\n",
      "1       1  0.998049  0.020202  0.999023  HD-SNN  0.328397\n",
      "2       1  0.998132  0.019358  0.999065  HD-SNN  0.336163\n",
      "3       1  0.998114  0.010989  0.999056  HD-SNN  0.337661\n",
      "4       1  0.998029  0.015906  0.999014  HD-SNN  0.348119\n",
      "..    ...       ...       ...       ...     ...       ...\n",
      "135   127  1.000000  1.000000  1.000000    corr  0.442675\n",
      "136   127  1.000000  1.000000  1.000000    corr  0.449080\n",
      "137   127  0.999999  0.999569  1.000000    corr  0.448772\n",
      "138   127  0.999999  0.999574  1.000000    corr  0.441505\n",
      "139   127  0.999999  0.999572  1.000000    corr  0.447610\n",
      "\n",
      "[140 rows x 6 columns]\n",
      "Bbox(x0=0.125, y0=0.25, x1=0.95, y1=0.975)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVMAAAFACAYAAAABCZ9BAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAAw3ElEQVR4nO3deVxUZd8G8It1ZHVBQUACF0xRBERBcu0Npce91SzDKJdUEiV9DFPRrDDNLbV40kTf3EhN60nDFCVzCRNEM7dEEEIWkRQFBGR+7x++To4sMnpgQK/v5zMfOffc9zm/mSMXZ5s5BiIiICKih2Ko7wKIiB4FDFMiIgUwTImIFMAwJSJSAMOUiEgBDFMiIgUwTImIFMAwJSJSAMOUiEgBDFMiIgXoNUz379+PQYMGwcHBAQYGBti+fft9x8TFxaFz585QqVRo06YN1qxZU+N1EhHdj17DtKCgAB4eHlixYkW1+qekpGDAgAF4+umnkZSUhEmTJmHUqFHYtWtXDVdKRFQ1g7ryRScGBgbYtm0bhg4dWmmfadOmYceOHTh58qSm7ZVXXsHVq1cRExNTC1USEVXMWN8F6OLw4cPw9/fXagsICMCkSZMqHVNcXIzi4mLNtFqtRl5eHmxsbGBgYFBTpRLRI0BEcP36dTg4OMDQsOod+XoVpllZWbCzs9Nqs7OzQ35+PoqKimBmZlZuTEREBObMmVNbJRLRIyg9PR0tWrSosk+9CtMHERYWhtDQUM30tWvX8MQTTyA9PR3W1tZ6rOzR80v/CK3pwCPLkVd6o9rjm5hY4n99gjXTPXeG1UhtE4+txoXCnGqPbWVui8+83qzxugBg4Znvse/KqWqPf9rGDe+2G1yn66rJ2vbn/IHf8/+q9lh36xboZduh2nXl5+fDyckJVlZW9513vQrT5s2bIzs7W6stOzsb1tbWFW6VAoBKpYJKpSrXbm1tzTBVmIWx9vsc1m4oTl/PqPb49laOWvNQcv3cPd+wdkORWphb7bEu5k1rpS4AmOD6LF5x7lHt8Tamlpp51NW6arK2fzl0xr8cOj/wvKpbV3UOCdarMPXz88POnTu12nbv3g0/Pz89VURV6dKkNbo0aa3vMsppY2WPNlb2+i6jQjYqK9io7r8VVNvqal11iV4vjbpx4waSkpKQlJQE4PalT0lJSUhLSwNwexc9MDBQ0//tt9/GhQsX8O9//xtnzpzB559/jm+++QaTJ0/WR/lERBp6DdOjR4/Cy8sLXl5eAIDQ0FB4eXlh1qxZAIDMzExNsAJAy5YtsWPHDuzevRseHh5YuHAhVq1ahYCAAL3UT0R0h1538/v06YOqLnOt6NNNffr0wbFjx2qwKiIi3fGz+URECmCYEhEpgGFKRKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQL0HqYrVqyAi4sLGjRoAF9fXxw5cqTK/kuWLMGTTz4JMzMzODk5YfLkybh582YtVUtEVDG9hml0dDRCQ0MRHh6OxMREeHh4ICAgADk5ORX237BhA9577z2Eh4fj9OnT+OqrrxAdHY3p06fXcuVERNr0GqaLFi3C6NGjERQUBDc3N0RGRsLc3ByrV6+usP+hQ4fQvXt3vPrqq3BxcUG/fv0wfPjw+27NEhHVNL2FaUlJCRISEuDv7/9PMYaG8Pf3x+HDhysc89RTTyEhIUETnhcuXMDOnTvRv3//SpdTXFyM/Px8rQcRkdKM9bXg3NxclJWVwc7OTqvdzs4OZ86cqXDMq6++itzcXPTo0QMiglu3buHtt9+ucjc/IiICc+bMUbR2IqJ76f0ElC7i4uLw8ccf4/PPP0diYiK+/fZb7NixA3Pnzq10TFhYGK5du6Z5pKen12LFRPS40NuWadOmTWFkZITs7Gyt9uzsbDRv3rzCMTNnzsTrr7+OUaNGAQDc3d1RUFCAMWPG4P3334ehYfm/DSqVCiqVSvkXQER0F71tmZqamsLb2xuxsbGaNrVajdjYWPj5+VU4prCwsFxgGhkZAQBEpOaKJSK6D71tmQJAaGgoRo4ciS5dusDHxwdLlixBQUEBgoKCAACBgYFwdHREREQEAGDQoEFYtGgRvLy84Ovri/Pnz2PmzJkYNGiQJlSJiPRBr2E6bNgwXL58GbNmzUJWVhY8PT0RExOjOSmVlpamtSU6Y8YMGBgYYMaMGcjIyECzZs0waNAgfPTRR/p6CUREAPQcpgAQHByM4ODgCp+Li4vTmjY2NkZ4eDjCw8NroTIiouqrV2fziYjqKoYpEZECGKZERApgmBIRKYBhSkSkAIYpEZECGKZERApgmBIRKYBhSkSkAIYpEZEC9P5xUnp4mZmZyMzMrHZ/e3t72Nvb12BFRI8fhukj4D//+Y9OdxMIDw/H7Nmza64goscQw/QRMHbsWAwePFgzXVRUhB49egAADhw4ADMzM63+3ColUh7D9BFw7257QUGB5mdPT09YWFjooyyixwrDtB663x56Sck/P3/0EWBq+uDzIqLqYZg+Aq5fz8SNG/+cgCotLdL8nJWVBBMT7d18S0t7WFlxV59ISQzTR0BCwn/w888Vn4CKiupRrq1373D06TO7hqsierwwTB8B3t5j8eSTg+/f8f9ZWnKrlEhpDNNHgJUVd9uJ9I2fgCIiUgDDlIhIAQxTIiIFMEyJiBTAMCUiUgDDlIhIAQxTIiIFMEyJiBTAMCUiUgDDlIhIAQxTIiIFMEyJiBTAMCUiUgDDlIhIAQxTIiIFMEyJiBSgc5i6uLjggw8+QFpaWk3UQ0RUL+kcppMmTcK3336LVq1aoW/fvti0aROKi4sfuIAVK1bAxcUFDRo0gK+vL44cOVJl/6tXr2LChAmwt7eHSqVC27ZtsXPnzgdePhGREh4oTJOSknDkyBG0b98e77zzDuzt7REcHIzExESd5hUdHY3Q0FCEh4cjMTERHh4eCAgIQE5OToX9S0pK0LdvX6SmpmLLli04e/YsVq5cCUdHR11fBhGRoh74mGnnzp3x2Wef4dKlSwgPD8eqVavQtWtXeHp6YvXq1RCR+85j0aJFGD16NIKCguDm5obIyEiYm5tj9erVFfZfvXo18vLysH37dnTv3h0uLi7o3bs3PDw8HvRlEBEp4oHDtLS0FN988w0GDx6Md999F126dMGqVavwwgsvYPr06XjttdeqHF9SUoKEhAT4+/v/U4yhIfz9/XH48OEKx3z//ffw8/PDhAkTYGdnh44dO+Ljjz9GWVnZg74MIiJF6Hx30sTERERFRWHjxo0wNDREYGAgFi9ejHbt2mn6PPfcc+jatWuV88nNzUVZWRns7Oy02u3s7HDmzJkKx1y4cAF79+7Fa6+9hp07d+L8+fMYP348SktLER4eXuGY4uJirWO6+fn51X2pRETVpnOYdu3aFX379sUXX3yBoUOHwsTEpFyfli1b4pVXXlGkwLup1WrY2triyy+/hJGREby9vZGRkYEFCxZUGqYRERGYM2eO4rUQEd1N5zC9cOECnJ2dq+xjYWGBqKioKvs0bdoURkZGyM7O1mrPzs5G8+bNKxxjb28PExMTGBkZadrat2+PrKwslJSUwNTUtNyYsLAwhIaGaqbz8/Ph5ORUZW1ERLrS+ZhpTk4O4uPjy7XHx8fj6NGj1Z6PqakpvL29ERsbq2lTq9WIjY2Fn59fhWO6d++O8+fPQ61Wa9rOnTsHe3v7CoMUAFQqFaytrbUeRERK0zlMJ0yYgPT09HLtGRkZmDBhgk7zCg0NxcqVK7F27VqcPn0a48aNQ0FBAYKCggAAgYGBCAsL0/QfN24c8vLyEBISgnPnzmHHjh34+OOPdV4uEZHSdN7NP3XqFDp37lyu3cvLC6dOndJpXsOGDcPly5cxa9YsZGVlwdPTEzExMZqTUmlpaTA0/CfvnZycsGvXLkyePBmdOnWCo6MjQkJCMG3aNF1fBhGRonQOU5VKhezsbLRq1UqrPTMzE8bGOs8OwcHBCA4OrvC5uLi4cm1+fn749ddfdV4OEVFN0nk3v1+/fggLC8O1a9c0bVevXsX06dPRt29fRYsjIqovdN6U/PTTT9GrVy84OzvDy8sLAJCUlAQ7Ozt8/fXXihdIRFQf6Bymjo6OOHHiBNavX4/jx4/DzMwMQUFBGD58eIXXnBIRPQ50P8iJ29eRjhkzRulaiIjqrQcKU+D2Wf20tDSUlJRotQ8ePPihiyIiqm8e6BNQzz33HH7//XcYGBhovh3KwMAAAPilI0T0WNL5bH5ISAhatmyJnJwcmJub448//sD+/fvRpUuXCi9lIiJ6HOi8ZXr48GHs3bsXTZs2haGhIQwNDdGjRw9ERERg4sSJOHbsWE3USURUp+m8ZVpWVgYrKysAt7+s5NKlSwAAZ2dnnD17VtnqiIjqCZ23TDt27Ijjx4+jZcuW8PX1xfz582Fqaoovv/yy3KeiiIgeFzqH6YwZM1BQUAAA+OCDDzBw4ED07NkTNjY2iI6OVrxAIqL6QOcwDQgI0Pzcpk0bnDlzBnl5eWjcuLHmjD4R0eNGp2OmpaWlMDY2xsmTJ7XamzRpwiAloseaTmFqYmKCJ554gteSEhHdQ+ez+e+//z6mT5+OvLy8mqiHiKhe0vmY6fLly3H+/Hk4ODjA2dkZFhYWWs8nJiYqVhwRUX2hc5gOHTq0BsogIqrfdA7Tym6pTET0ONP5mCkREZWn85apoaFhlZdB8Uw/ET2OdA7Tbdu2aU2Xlpbi2LFjWLt2LebMmaNYYURE9YnOYTpkyJBybS+++CI6dOiA6OhovPXWW4oURkRUnyh2zLRbt26IjY1VanZERPWKImFaVFSEzz77DI6OjkrMjoio3tF5N//eLzQREVy/fh3m5uZYt26dosUREdUXOofp4sWLtcLU0NAQzZo1g6+vLxo3bqxocURE9YXOYfrGG2/UQBlERPWbzsdMo6KisHnz5nLtmzdvxtq1axUpioiovtE5TCMiItC0adNy7ba2tvj4448VKYqIqL7ROUzT0tLQsmXLcu3Ozs5IS0tTpCgiovpG5zC1tbXFiRMnyrUfP34cNjY2ihRFRFTf6Bymw4cPx8SJE7Fv3z6UlZWhrKwMe/fuRUhICF555ZWaqJGIqM7T+Wz+3LlzkZqaimeeeQbGxreHq9VqBAYG8pgpET22dA5TU1NTREdH48MPP0RSUhLMzMzg7u4OZ2fnmqiPiKhe0DlM73B1dYWrq6uStRAR1Vs6HzN94YUX8Mknn5Rrnz9/Pl566SVFiiIiqm90DtP9+/ejf//+5dr/9a9/Yf/+/Q9UxIoVK+Di4oIGDRrA19cXR44cqda4TZs2wcDAgPelIiK90zlMb9y4AVNT03LtJiYmyM/P17mA6OhohIaGIjw8HImJifDw8EBAQABycnKqHJeamoopU6agZ8+eOi+TiEhpOoepu7s7oqOjy7Vv2rQJbm5uOhewaNEijB49GkFBQXBzc0NkZCTMzc2xevXqSseUlZXhtddew5w5c9CqVSudl0lEpDSdT0DNnDkTzz//PJKTk/E///M/AIDY2Fhs2LABW7Zs0WleJSUlSEhIQFhYmKbN0NAQ/v7+OHz4cKXjPvjgA9ja2uKtt97CL7/8outLICJSnM5hOmjQIGzfvh0ff/wxtmzZAjMzM3h4eGDv3r1o0qSJTvPKzc1FWVkZ7OzstNrt7Oxw5syZCsccOHAAX331FZKSkqq1jOLiYhQXF2umH+RQBBHR/TzQN+0PGDAABw8eREFBAS5cuICXX34ZU6ZMgYeHh9L1abl+/Tpef/11rFy5ssIvW6lIREQEGjZsqHk4OTnVaI1E9Hh64OtM9+/fj6+++gpbt26Fg4MDnn/+eaxYsUKneTRt2hRGRkbIzs7Was/Ozkbz5s3L9U9OTkZqaioGDRqkaVOr1QAAY2NjnD17Fq1bt9YaExYWhtDQUM10fn4+A5WIFKdTmGZlZWHNmjX46quvkJ+fj5dffhnFxcXYvn37A518MjU1hbe3N2JjYzWXN6nVasTGxiI4OLhc/3bt2uH333/XapsxYwauX7+OpUuXVhiSKpUKKpVK59qIiHRR7TAdNGgQ9u/fjwEDBmDJkiV49tlnYWRkhMjIyIcqIDQ0FCNHjkSXLl3g4+ODJUuWoKCgAEFBQQCAwMBAODo6IiIiAg0aNEDHjh21xjdq1AgAyrUTEdWmaofpjz/+iIkTJ2LcuHGKfox02LBhuHz5MmbNmoWsrCx4enoiJiZGc1IqLS0NhoaK3ZGaiKhGVDtM75xF9/b2Rvv27fH6668r9pV7wcHBFe7WA0BcXFyVY9esWaNIDURED6Pam3zdunXDypUrkZmZibFjx2LTpk1wcHCAWq3G7t27cf369Zqsk4ioTtN5/9nCwgJvvvkmDhw4gN9//x3vvvsu5s2bB1tbWwwePLgmaiQiqvMe6mDkk08+ifnz5+Ovv/7Cxo0blaqJiKjeUeTMjpGREYYOHYrvv/9eidkREdU7PE1ORKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQIYpkRECmCYEhEpgGFKRKQAhikRkQKqfatnqruuX8/EjRuZ1e5vaWkPKyv7GqyI6PHDMH0EJCT8Bz//PKfa/Xv3DkefPrNrriCixxDD9BHg7T0WTz75z222S0uLEBXVAwAQFHQAJiZmWv0tLblVSqQ0hukjwMpKe7e9pKRA83Pz5p4wNbXQR1lEjxWegCIiUgC3TOuhPnGzq3y+qKwEEf//c89fPoKZkWkVvaueFxFVD7dMiYgUwDAlIlIAw5SISAE8ZvoIuFJ8HVdKbmimi9Wlmp/P38iCytBEq7+NqSVsVFa1Vh/R44Bh+gj4b2YC1l78ucLnJiZFlWsb6dwbb7j0qeGqiB4vDNNHwCB7bzxl82S1+9uYWtZgNUSPJ4bpI8BGZcXddiI94wkoIiIFMEyJiBRQJ8J0xYoVcHFxQYMGDeDr64sjR45U2nflypXo2bMnGjdujMaNG8Pf37/K/kREtUHvYRodHY3Q0FCEh4cjMTERHh4eCAgIQE5OToX94+LiMHz4cOzbtw+HDx+Gk5MT+vXrh4yMjFqunIjoH3oP00WLFmH06NEICgqCm5sbIiMjYW5ujtWrV1fYf/369Rg/fjw8PT3Rrl07rFq1Cmq1GrGxsbVcORHRP/QapiUlJUhISIC/v7+mzdDQEP7+/jh8+HC15lFYWIjS0lI0adKkwueLi4uRn5+v9SAiUppewzQ3NxdlZWWws7PTarezs0NWVla15jFt2jQ4ODhoBfLdIiIi0LBhQ83DycnpoesmIrqX3nfzH8a8efOwadMmbNu2DQ0aNKiwT1hYGK5du6Z5pKen13KVRPQ40OtF+02bNoWRkRGys7O12rOzs9G8efMqx3766aeYN28e9uzZg06dOlXaT6VSQaVSKVIvEVFl9LplampqCm9vb62TR3dOJvn5+VU6bv78+Zg7dy5iYmLQpUuX2iiViKhKev84aWhoKEaOHIkuXbrAx8cHS5YsQUFBAYKCggAAgYGBcHR0RETE7e+O/+STTzBr1ixs2LABLi4ummOrlpaWsLTkZ86JSD/0HqbDhg3D5cuXMWvWLGRlZcHT0xMxMTGak1JpaWkwNPxnA/qLL75ASUkJXnzxRa35hIeHY/bs2bVZOhGRht7DFACCg4MRHBxc4XNxcXFa06mpqTVfEBGRjur12XwiorqCYUpEpACGKRGRAhimREQKYJgSESmAYUpEpACGKRGRAhimREQKYJgSESmAYUpEpACGKRGRAhimREQKYJgSESmAYUpEpACGKRGRAhimREQKYJgSESmAYUpEpACGKRGRAhimREQKYJgSESmAYUpEpACGKRGRAhimREQKYJgSESmAYUpEpACGKRGRAhimREQKYJgSESmAYUpEpACGKRGRAhimREQKYJgSESmAYUpEpABjfRdQX2RmZiIzM7Pa/e3t7WFvb1+DFRFRXcIwrab//Oc/mDNnTrX7h4eHY/bs2TVXEBHVKXUiTFesWIEFCxYgKysLHh4eWLZsGXx8fCrtv3nzZsycOROpqalwdXXFJ598gv79+yte191ZmJzcHU8/PVczfetWKX755QMAQM+es2BsbKI1NjnZV2s8c5Xo0ab3MI2OjkZoaCgiIyPh6+uLJUuWICAgAGfPnoWtrW25/ocOHcLw4cMRERGBgQMHYsOGDRg6dCgSExPRsWPHGqtz375ZyMj4tcLn7oTq3Rwdu6F16741Vg8R1S16D9NFixZh9OjRCAoKAgBERkZix44dWL16Nd57771y/ZcuXYpnn30WU6dOBQDMnTsXu3fvxvLlyxEZGalobX3iZmt+TigrQYYOY73KSrTGA7Mr6UlEjwK9hmlJSQkSEhIQFhamaTM0NIS/vz8OHz5c4ZjDhw8jNDRUqy0gIADbt2+vsH9xcTGKi4s109euXQMA5Ofn37e+glv/jHu5hR/87Trdd8wdTUwstMZXZ3nVdfd8H1ZdrQuou7WxLt3V1druV9ed50Xk/jMTPcrIyBAAcujQIa32qVOnio+PT4VjTExMZMOGDVptK1asEFtb2wr7h4eHCwA++OCDjwd+pKen3zfP9L6bX9PCwsK0tmTVajXy8vJgY2MDAwODh5p3fn4+nJyckJ6eDmtr64ctVVF1tTbWpbu6WtvjUJeI4Pr163BwcLhvX72GadOmTWFkZITs7Gyt9uzsbDRv3rzCMc2bN9epv0qlgkql0mpr1KjRgxddAWtr6zr1n+ludbU21qW7ulrbo15Xw4YNq9VPr5+AMjU1hbe3N2JjYzVtarUasbGx8PPzq3CMn5+fVn8A2L17d6X9iYhqg95380NDQzFy5Eh06dIFPj4+WLJkCQoKCjRn9wMDA+Ho6IiIiAgAQEhICHr37o2FCxdiwIAB2LRpE44ePYovv/xSny+DiB5zeg/TYcOG4fLly5g1axaysrLg6emJmJgY2NnZAQDS0tJgaPjPBvRTTz2FDRs2YMaMGZg+fTpcXV2xffv2Gr3GtDIqlQrh4eHlDiPUBXW1Ntalu7paG+vSZiBSnXP+RERUFX5rFBGRAhimREQKYJgSESmAYVoDbt26pe8SKlRYWKjvEiqVmpqq7xK0cB3Wf7W9DhmmChIRxMfHY+XKlcjNzdV3ORoign379mHq1KnIyNDl61pqR0pKClq1aoWzZ8/quxSuQx3d/b0XdYW+1iHDVEEGBgbIycnBd999hylTpui7HA0DAwM4OTkhLy8PY8eO1Xc5WkQELVu2xBtvvIFRo0bpuxyuw2q6efMmZs6cicGDB8Pd3R1BQUH48ccf9V0WAD2uw/t+ep+qpbi4WPNzdHS0GBgYyM8//6zHim4rKSnR/Jyeni7NmjWTnTt36rGiiuXl5YmFhYVs375dbzVwHd6fWq2WdevWiaGhoVhYWMiMGTNk48aNEhoaKq1atdL7/y19rkOG6UMqLS3Vmp47d66Ym5tLp06dJCYmptbrycvL0/xcWloq06dPl3379mnaQkJC5MUXX6z1ukRu/yLe/e+9li5dKi1atKjNkkSE67C6bty4IS+++KI0a9ZMPD09pWfPntK3b185ePCgiIh8+OGH0qpVq1qvS6RurEOG6QNISUmR8+fPa7Vt3rxZHBwcpFmzZhIRESEpKSnlVnBNUqvVsnjxYpk8ebLWlszkyZOldevWmukXXnhBJkyYUGmg1abk5GS5cOGC/Pjjj/Lpp5/KjBkzxMDAQNasWVPjy+Y61N3ixYvFyclJVq1aJYWFhVJYWCjLli0TR0dHKSkpkWvXrombm5vs2LGjVuqpzjq8cOFCra1DhukDCA8PF3d3dxEROXnypPj5+YmxsbGMGTNGjh07prWrUZtmz54tffv2FZHbv5hlZWVy7tw5MTc3l+eff17c3d3FwsKi3PfB1obc3FxZs2aNTJ06VSZMmCBeXl7i5uYmtra20qBBA+nQoYO4u7vLG2+8IYsWLarxeur7Ovz6669rta7S0lIZMmSIBAcHa7VnZGRIt27d5ODBg1JaWip79uyRtLS0Wqmprq1DhukDKCwsFHNzc+nSpYsYGRlJ37595aeffpL8/Hy91lVUVCTW1tayZcsWTVtSUpK4u7vLrl27JCQkRKKjo/USFCUlJdK0aVPp2rWr9OrVS4KDg+Xzzz+XjRs3SnZ2drktjJqmyzosLS2V33//vVbqut86nDhxokRHR4uIyK1bt+TSpUu1UtetW7fE29tbli9frtUeGRkpjRo1kj///FNEbr+vtUWXdZiXlycpKSk1Wg/D9AGtWrVKDAwMZOnSpZKdna3vcjSWLVsmHTt2lJUrV8qOHTvklVdekV69esnNmzf1XZpcu3ZNbt68qbULe6/a3HWtzjo8ePCgTJgwQVq1aiWnTp2qlbqquw6XLVsm48ePl4sXL9ZKXQsXLhRXV1eJj4+XpKQk2bBhg/Tp00dGjx5d7g/08ePH5cqVKzVeU3XW4d9//y39+vWTgICAGv2jzTB9CM7OzjJx4kR9l1HORx99JO3atRM7Ozvp3r27bN68Wd8llXPr1i0Rqd3wrMi96/BOPZcvX5ZPPvlEunfvLv379xdXV1fx8vKqtboqW4dqtVpTY1xcnAwcOFAGDx5ca3WNGTNGXF1dpXXr1uLi4iIDBw6UCxcuaPU5e/asBAcHy5tvvlkrNT3xxBOV/h7eea+OHDkib731lvTq1avG6mCYPoSYmBjp2rWrZGVl6buUcjIyMuTEiRO1tjVVX8XExEjnzp3l8uXLWu3JyckSEhIiDRs2lJMnT4qIiI2Njaxbt67WaktLS6tyHebm5kpYWJgYGBhIbGxsrdRUVFQkZ86ckT179siBAwcq7KNWqyUhIUEcHBxk06ZNNV5TTEyMeHt7S25uriY87/yxvltWVpa0aNFCfvjhhxqpg2H6kBITE/VdAj2kU6dOyfXr1+Xo0aNa7RcuXBB3d3dZuXKliNw+m21jY6OPErUUFRXJnj17pE+fPqJSqWT8+PFSUFCgt3runC3PycmRzMxMTXtkZKS0adOmVmpISkqS7OxsOXfuXLnn7mzNl5aWytChQ2XGjBk1UgM/AfWQvLy89F0CPSRXV1eMGzcOa9euhVqt1rS3bNkSo0ePxrp161BSUoJJkyahb9++ld6GXGm3bt3C8uXLER8fD+D2LX1OnjyJ8ePHIyAgAMbGxjh27BhWrFgBc3PzWqnpbnv27EFmZiaMjW9/x3xiYiL69++veb5nz55QqVQ4evRojdfSpk0bdOrUCe7u7nj22Wfx/vvvY8+ePbhy5Yrmxpk5OTk4evQo2rZtWyM1MEzpsWdsbIyWLVsiMTFRc1eHsrIyAMClS5dgZGSkuW96VFRUrd1vzNjYGH/99RdGjBiBc+fOYd68eejVqxf279+Pbdu2Yffu3Wjfvn317umusL179+Lf//43Ll++rGnr3r07MjMz8d5776G0tBSHDh3ClStX4OjoWOP1WFhYICgoCGq1Gi+++CLi4+Px1ltvwdfXF76+vnjppZfg4eEBKysr9OrVq2aKqJHtXaJ6Rq1WS5MmTWTVqlVy48YNEbl9TLJjx47y0ksv6bUuGxsbsbKykoYNG8onn3xS7nl9adOmjaxdu1ZE/vnI65gxY8Td3V1at24tBgYGMm3atFqrR61Wi7m5uXz33XcicvsY6YEDB2ThwoUyatQoWbRokWbdqtVqxS9lZJgS/b/169eLu7u7PPPMMzJq1Cixs7MTHx8fSU9P12tdGzZsEAMDAzlx4oSmraysTI8V3RYZGSktW7bUun7Tz89PVqxYIUePHpVTp05JUVFRrda0fPlycXBwqPLyO5Hbn+EPCgqq9CTag+A9oIjusnfvXkRFRSE5ORnPPvsshgwZAg8PD4iI5tibPjzxxBPo378/vvjiC73Wca+BAweiuLgYLVu2RHp6OlJTU7Ft2za0a9dObzU5ODhgxIgRmD9/vla7Wq3WujnnxIkTsXPnTpw/f16R5TJMiSpx7y+fPsXHx2Pjxo348MMPYWlpqe9yNDIyMvD999/js88+Q6dOndC3b1+8+eabMDAw0Fvox8fH48CBA5g0aRKMjIwq7Xfz5k20bNkSs2fPVuRrDRmmRPfQ91ZoZa5evYpGjRrpu4wK3bp1C6WlpTAzM9N3KQCA0tJSmJiYaKbvrNOysjIYGRlp/lBGR0djzJgxuHbt2kMvs2782SWqQ+pikAKos0EK3L7yoK4EKQCtIAX+Wad3tlQLCgogImjVqhWMjY0xderUh16m8UPPgYiojkpOToaNjQ1OnDiB8+fPIzc3Fz///DPKyspw6NAh2NraoqCgAAcOHEBRUdFD/UHgbj4RPbJefvll7Nq1C2q1Gg4ODigoKECPHj1gZWUFX19fuLi4oGnTpmjfvj1MTEwe6hg5w5SIHllHjhxBYmIinnnmGRQVFaFTp041tiyGKRE9Vu6cfFL6ag2GKRGRAng2n4hIAQxTIiIFMEyJiBTAMCUiUgDDlIhIAQxTIiIFMEyJiBTAMCUiUgDDlIhIAQxTIiIFMExJS58+fTBp0iR9l6GxZs0aRb7HszrzmT17Njw9PTXTb7zxBoYOHaqZ1ud7c/DgQbi7u8PExESrJn2p6P388ssv4eTkBENDQyxZskQvdekTw5TqtGHDhuHcuXOa6XsDT0lTpkxBbGxspc9/++23mDt3rmbaxcWl1kIjNDQUnp6eSElJwZo1a2plmXdU9DrvXS/5+fkIDg7GtGnTkJGRgTFjxtRqjXUBvxya6jQzM7Na+wZ3S0vLKu+v1KRJk1qpoyLJycl4++230aJFC73VcLd710taWhpKS0sxYMAA2Nvb67Ey/eGWaT3Rp08fBAcHIzg4GA0bNkTTpk0xc+ZM3P2lX8XFxZgyZQocHR1hYWEBX19fxMXFaZ6/cuUKhg8fDkdHR5ibm8Pd3R0bN26scrk7duxAw4YNsX79+kr7nDx5Ev/6179gaWkJOzs7vP7668jNzQUAxMXFwdTUFL/88oum//z582Fra4vs7GwAt+9tNHbsWNjZ2aFBgwbo2LEjfvjhBwDau5Nr1qzBnDlzcPz4cc0N2+5spS1atAju7u6wsLCAk5MTxo8fjxs3bpSrdfv27XB1dUWDBg0QEBCA9PR0zXP32+q9eze/T58+uHjxIiZPnqyppaCgANbW1tiyZUu5ZVpYWOD69esVzre4uBgTJ06Era0tGjRogB49euC3334DAKSmpsLAwABXrlzR3Kiusi1TFxcXfPjhhwgMDISlpSWcnZ3x/fff4/LlyxgyZAgsLS3RqVMnHD16VGvc1q1b0aFDB6hUKri4uGDhwoVar/ne13lnXdy9Xtzd3QEArVq1goGBAVJTU3H8+HE8/fTTsLKygrW1Nby9vcst+5Gi2E2jqUb17t1bLC0tJSQkRM6cOSPr1q0Tc3Nz+fLLLzV9Ro0aJU899ZTs379fzp8/LwsWLBCVSiXnzp0TEZG//vpLFixYIMeOHZPk5GT57LPPxMjISOLj47WWExISIiK37yNvZWUl//3vfyut6++//5ZmzZpJWFiYnD59WhITE6Vv377y9NNPa/pMnTpVnJ2d5erVq5KYmCimpqby3Xfficjt+79369ZNOnToID/99JMkJyfLf//7X9m5c6eIiERFRUnDhg1FRKSwsFDeffdd6dChg2RmZkpmZqYUFhaKiMjixYtl7969kpKSIrGxsfLkk0/KuHHjNDVERUWJiYmJdOnSRQ4dOiRHjx4VHx8feeqppzR9wsPDxcPDQzM9cuRIGTJkSIXvzZUrV6RFixbywQcfaGoRERk9erT0799f6z0aPHiwBAYGVvoeTpw4URwcHGTnzp3yxx9/yMiRI6Vx48Zy5coVuXXrlmRmZoq1tbUsWbJE6zXfy9nZWZo0aSKRkZFy7tw5GTdunFhbW8uzzz4r33zzjZw9e1aGDh0q7du3F7VaLSIiR48eFUNDQ/nggw/k7NmzEhUVJWZmZhIVFVXl67x3vezZs0cAyJEjRyQzM1Nu3bolHTp0kBEjRsjp06fl3Llz8s0330hSUlKl70N9xzCtJ3r37q31SyAiMm3aNGnfvr2IiFy8eFGMjIwkIyNDa9wzzzwjYWFhlc53wIAB8u6772otJyQkRJYvXy4NGzaUuLi4KuuaO3eu9OvXT6stPT1dAMjZs2dFRKS4uFg8PT3l5ZdfFjc3Nxk9erSm765du8TQ0FDT9153/9KKlA+8ymzevFlsbGy05gNAfv31V03b6dOnBYDmj4kuYSpyO7wWL16stdz4+HgxMjKSS5cuiYhIdna2GBsbV/o+3rhxQ0xMTGT9+vWatpKSEnFwcJD58+dr2ho2bKgJuMo4OzvLiBEjNNOZmZkCQGbOnKlpO3z4sADQhOKrr74qffv21ZrP1KlTxc3NrcrXee96OXbsmACQlJQUTZuVlZWsWbOmypofJdzNr0e6deumdedMPz8//PnnnygrK8Pvv/+OsrIytG3bVnPsz9LSEj///DOSk5MBAGVlZZg7dy7c3d3RpEkTWFpaYteuXUhLS9NazpYtWzB58mTs3r0bvXv3rrKm48ePY9++fVrLbNeuHQBolmtqaor169dj69atuHnzJhYvXqwZn5SUhBYtWqBt27YP9d7s2bMHzzzzDBwdHWFlZYXXX38dV65cQWFhoaaPsbExunbtqplu164dGjVqhNOnTz/Usu/m4+ODDh06YO3atQCAdevWwdnZGb169aqwf3JyMkpLS9G9e3dNm4mJCXx8fB6orrtvy2FnZwcAml3wu9tycnIAAKdPn9ZaNgB0795d8//qYYSGhmLUqFHw9/fHvHnzNP8fHlUM00fEjRs3YGRkhISEBCQlJWkep0+fxtKlSwEACxYswNKlSzFt2jTs27cPSUlJCAgIQElJida8vLy80KxZM6xevVrrmGxlyx00aJDWMpOSkvDnn39qBcihQ4cAAHl5ecjLy9O0K3FyKTU1FQMHDkSnTp2wdetWJCQkYMWKFQBQ7rXVhlGjRmmOa0ZFRSEoKKjWbh999y2O7yyzoja1Wl3jtcyePRt//PEHBgwYgL1798LNzQ3btm2r8eXqC8O0HomPj9ea/vXXX+Hq6gojIyN4eXmhrKwMOTk5aNOmjdajefPmAG5fqzhkyBCMGDECHh4eaNWqldblLXe0bt0a+/btw3fffYd33nmnypo6d+6MP/74Ay4uLuWWa2FhAeD21tfkyZOxcuVK+Pr6YuTIkZpf5k6dOuGvv/6qsI6KmJqalttiSkhIgFqtxsKFC9GtWze0bdsWly5dKjf21q1bWidAzp49i6tXr6J9+/bVWnZ1agGAESNG4OLFi/jss89w6tQpjBw5stJ5tG7dGqampjh48KCmrbS0FL/99hvc3NweqC5dtG/fXmvZwO3/J23bttXcY76y11kdbdu2xeTJk/HTTz/h+eefR1RU1EPXXFcxTOuRtLQ0hIaG4uzZs9i4cSOWLVuGkJAQALf/07722msIDAzEt99+i5SUFBw5cgQRERHYsWMHAMDV1RW7d+/GoUOHcPr0aYwdO1ZzRv1ebdu2xb59+7B169YqL1SfMGEC8vLyMHz4cPz2229ITk7Grl27EBQUhLKyMpSVlWHEiBEICAhAUFAQoqKicOLECc0Z4969e6NXr1544YUXsHv3bqSkpODHH39ETExMhctzcXFBSkoKkpKSkJubi+LiYrRp0walpaVYtmwZLly4gK+//hqRkZHlxpqYmOCdd95BfHw8EhIS8MYbb6Bbt27w8fHRZTVo1bJ//35kZGRorl4AgMaNG+P555/H1KlT0a9fvyovZ7KwsMC4ceMwdepUxMTE4NSpUxg9ejQKCwvx1ltvPVBdunj33XcRGxuLuXPn4ty5c1i7di2WL1+OKVOmaPpU9jqrUlRUhODgYMTFxeHixYs4ePAgfvvttwf+w1Uv6PugLVVP7969Zfz48fL222+LtbW1NG7cWKZPn651QqqkpERmzZolLi4uYmJiIvb29vLcc8/JiRMnROT2mdkhQ4aIpaWl2NrayowZMyQwMLDKkyynTp0SW1tbCQ0NrbS2c+fOyXPPPSeNGjUSMzMzadeunUyaNEnUarXMmTNH7O3tJTc3V9N/69atYmpqqjmze+XKFQkKChIbGxtp0KCBdOzYUX744QcRKX+i4+bNm/LCCy9Io0aNBIDmpMyiRYvE3t5ezMzMJCAgQP73f/9XAMjff/+tNZ+tW7dKq1atRKVSib+/v1y8eFEzb11PQB0+fFg6deokKpVK7v1Vio2NFQDyzTffVPq+3VFUVCTvvPOONG3aVFQqlXTv3l2OHDmi1ae6J6DuPVEEQLZt26aZTklJEQBy7NgxTduWLVvEzc1NTExM5IknnpAFCxZozaOi13m/E1DFxcXyyiuviJOTk5iamoqDg4MEBwdLUVHRfd+P+op3J60n+vTpA09Pz8fyY3r10ddff43Jkyfj0qVLMDU11Xc5VAv4CSgiBRUWFiIzMxPz5s3D2LFjGaSPER4zJVLQ/Pnz0a5dOzRv3hxhYWH6LodqEXfziYgUwC1TIiIFMEyJiBTAMCUiUgDDlIhIAQxTIiIFMEyJiBTAMCUiUgDDlIhIAQxTIiIF/B8Fo3cwhxDRdAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 325x325 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading filename='data_cache/2023-09-27_scan_tau_decay.json'...\n",
      "     tau_decay  accuracy        TP        TN  method      time\n",
      "0          0.1  0.998340  0.129670  0.999169  HD-SNN  0.378289\n",
      "1          0.1  0.998241  0.116582  0.999119  HD-SNN  0.385068\n",
      "2          0.1  0.998360  0.138583  0.999180  HD-SNN  0.385842\n",
      "3          0.1  0.998337  0.127912  0.999168  HD-SNN  0.381018\n",
      "4          0.1  0.998289  0.145668  0.999144  HD-SNN  0.381509\n",
      "..         ...       ...       ...       ...     ...       ...\n",
      "175        0.9  0.999955  0.976684  0.999977    corr  0.443320\n",
      "176        0.9  0.999947  0.972173  0.999974    corr  0.442818\n",
      "177        0.9  0.999941  0.970289  0.999971    corr  0.445000\n",
      "178        0.9  0.999975  0.986882  0.999987    corr  0.445637\n",
      "179        0.9  0.999962  0.980511  0.999981    corr  0.442980\n",
      "\n",
      "[180 rows x 6 columns]\n",
      "Bbox(x0=0.125, y0=0.25, x1=0.95, y1=0.975)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x325 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading filename='data_cache/2023-09-27_scan_tau_rise.json'...\n",
      "     tau_rise  accuracy        TP        TN  method      time\n",
      "0        0.01  0.999975  0.986813  0.999987  HD-SNN  0.443718\n",
      "1        0.01  0.999967  0.983586  0.999984  HD-SNN  0.442027\n",
      "2        0.01  0.999978  0.988561  0.999989  HD-SNN  0.442102\n",
      "3        0.01  0.999982  0.990769  0.999991  HD-SNN  0.445676\n",
      "4        0.01  0.999974  0.987024  0.999987  HD-SNN  0.445614\n",
      "..        ...       ...       ...       ...     ...       ...\n",
      "115      0.90  0.998065  0.005172  0.999032    corr  0.318829\n",
      "116      0.90  0.998022  0.007569  0.999010    corr  0.322723\n",
      "117      0.90  0.998081  0.005215  0.999040    corr  0.309788\n",
      "118      0.90  0.998093  0.009146  0.999046    corr  0.308627\n",
      "119      0.90  0.998077  0.008643  0.999038    corr  0.316173\n",
      "\n",
      "[120 rows x 6 columns]\n",
      "Bbox(x0=0.125, y0=0.25, x1=0.95, y1=0.975)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x325 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading filename='data_cache/2023-09-27_scan_p_SM.json'...\n",
      "      p_SM  accuracy        TP        TN  method      time\n",
      "0    0.005  0.999385  0.677363  0.999692  HD-SNN  0.249565\n",
      "1    0.005  0.999379  0.688131  0.999689  HD-SNN  0.247990\n",
      "2    0.005  0.999416  0.693357  0.999708  HD-SNN  0.258380\n",
      "3    0.005  0.999425  0.698022  0.999712  HD-SNN  0.256826\n",
      "4    0.005  0.999392  0.696526  0.999696  HD-SNN  0.259028\n",
      "..     ...       ...       ...       ...     ...       ...\n",
      "235  1.000  0.999815  0.904987  0.999907    corr  0.440625\n",
      "236  1.000  0.999776  0.883406  0.999888    corr  0.447725\n",
      "237  1.000  0.999769  0.881647  0.999884    corr  0.442866\n",
      "238  1.000  0.999800  0.893966  0.999900    corr  0.447613\n",
      "239  1.000  0.999764  0.877338  0.999882    corr  0.446120\n",
      "\n",
      "[240 rows x 6 columns]\n",
      "Bbox(x0=0.125, y0=0.25, x1=0.95, y1=0.975)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x325 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading filename='data_cache/2023-09-27_scan_p_A.json'...\n",
      "       p_A  accuracy        TP        TN  method      time\n",
      "0    0.001  0.999881  0.937582  0.999940  HD-SNN  0.446202\n",
      "1    0.001  0.999896  0.947811  0.999948  HD-SNN  0.453300\n",
      "2    0.001  0.999908  0.951606  0.999954  HD-SNN  0.445786\n",
      "3    0.001  0.999910  0.952967  0.999955  HD-SNN  0.448156\n",
      "4    0.001  0.999912  0.956049  0.999956  HD-SNN  0.446265\n",
      "..     ...       ...       ...       ...     ...       ...\n",
      "155  0.500  0.998546  0.253230  0.999273    corr  0.437053\n",
      "156  0.500  0.998778  0.371552  0.999388    corr  0.441794\n",
      "157  0.500  0.998800  0.392706  0.999399    corr  0.440919\n",
      "158  0.500  0.998886  0.427895  0.999442    corr  0.440956\n",
      "159  0.500  0.998717  0.323607  0.999358    corr  0.440026\n",
      "\n",
      "[160 rows x 6 columns]\n",
      "Bbox(x0=0.125, y0=0.25, x1=0.95, y1=0.975)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x325 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading filename='data_cache/2023-09-27_scan_p_B.json'...\n",
      "       p_B  accuracy        TP        TN  method      time\n",
      "0    0.001  0.999881  0.937582  0.999940  HD-SNN  0.442927\n",
      "1    0.001  0.999896  0.947811  0.999948  HD-SNN  0.441032\n",
      "2    0.001  0.999908  0.951606  0.999954  HD-SNN  0.447361\n",
      "3    0.001  0.999910  0.952967  0.999955  HD-SNN  0.443855\n",
      "4    0.001  0.999912  0.956049  0.999956  HD-SNN  0.452339\n",
      "..     ...       ...       ...       ...     ...       ...\n",
      "155  0.500  0.606247  0.594893  0.616983    corr  0.460322\n",
      "156  0.500  0.605156  0.593365  0.616283    corr  0.447884\n",
      "157  0.500  0.606234  0.594219  0.617558    corr  0.453909\n",
      "158  0.500  0.603978  0.591834  0.615421    corr  0.450483\n",
      "159  0.500  0.605717  0.593583  0.617148    corr  0.449912\n",
      "\n",
      "[160 rows x 6 columns]\n",
      "Bbox(x0=0.125, y0=0.25, x1=0.95, y1=0.975)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVMAAAFLCAYAAABrzm+CAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAA6yklEQVR4nO3deVxU5f4H8A8gDDspyKYIml5ERBAQxA0rilte98qtRDK03DBalDLRuoXmEi2U5YL+yjW9WqbhgpKKpjcQzQUVlyADVFBZVMDh+f3hZWRkWAbPMAf8vF+veTnzzHPO+cyZ4evZj4EQQoCIiB6Kob4DEBE1ByymREQSYDElIpIAiykRkQRYTImIJMBiSkQkARZTIiIJsJgSEUmAxZSISAIspkREEtBrMd23bx8GDhwIZ2dnGBgYYMuWLXUOk5ycDF9fXygUCnTs2BErV67UeU4iorrotZiWlJTA29sb8fHx9ep/8eJFDBgwAE888QTS09Mxffp0vPrqq9ixY4eOkxIR1c5ALhc6MTAwwObNmzFkyJAa+8yYMQPbtm3DiRMnVG0jR47EjRs3kJiY2AgpiYg0a6HvANo4dOgQQkJC1NpCQ0Mxffr0GocpLS1FaWmp6nVFRQUKCgpga2sLAwMDXUUlomZACIGioiI4OzvD0LD2FfkmVUxzc3Ph4OCg1ubg4IDCwkLcvn0bZmZm1YaJjY3F3LlzGysiETVD2dnZaNu2ba19mlQxbYjo6GhERUWpXt+8eRPt2rVDdnY2rK2t9ZiMiOSusLAQLi4usLKyqrNvkyqmjo6OyMvLU2vLy8uDtbW1xqVSAFAoFFAoFNXara2tWUyJqF7qs0mwSR1nGhQUhKSkJLW2Xbt2ISgoSE+JiIju0WsxLS4uRnp6OtLT0wHcO/QpPT0dWVlZAO6too8dO1bV/7XXXsOFCxfwzjvvICMjA1999RU2bNiAN954Qx/xiYhU9FpMf//9d3Tv3h3du3cHAERFRaF79+6YPXs2ACAnJ0dVWAGgffv22LZtG3bt2gVvb28sWrQIy5YtQ2hoqF7yExFVks1xpo2lsLAQNjY2uHnzJreZElGttKkXTWqbKRGRXDWpvfmPspycHOTk5NS7v5OTE5ycnHSYiIiqYjGtgz6L2Jw595/v2LEQv/22uN7D9uwZhdDQRWrjICLdYTGtwzfffKPVGVQxMTGY00wrGJeOiWrGYlqHiRMnYtCgQarXt2/fRp8+fQAABw4cqHaygK6KR69eb6FbtzH17m9pKV2Oyv8buHRMVDMW0xpUFoCiIqC4+H57efn958uXA8bG6sNZWgJWVpC8gFhZOcHKikt5RHLFYlqH1NRv8OuvmlfzExL6VGsLDo5B//5zdJxKP/S5dEwkdyymdfDzmwh390F1d/yf5lxAuHRMVDMW0zqwgBBRffCgfSIiCbCYEhFJgMWUiEgCLKZERBJgMSUikgCLKRGRBHhoFDU5vEYAyRGLKTUJvIIWyR1X84mIJMAlU2pyeI0AkiMWU2pyeIovyRFX84mIJMBiSkQkARZTIiIJsJgSEUmAxZSISAIspkREEmAxJSKSAIspEZEEWEyJiCTAYkpEJAEWUyIiCbCYEhFJgMWUiEgCLKZERBJgMSUikgCLKRGRBPReTOPj4+Hm5gZTU1MEBgbiyJEjtfaPi4uDu7s7zMzM4OLigjfeeAN37txppLRERJrptZiuX78eUVFRiImJQVpaGry9vREaGoorV65o7L9mzRrMnDkTMTExOH36NJYvX47169fj3XffbeTkRETq9FpMFy9ejIiICISHh6NLly5YsmQJzM3NsWLFCo39Dx48iN69e2P06NFwc3PDM888g1GjRtW5NEtEpGt6K6ZlZWVITU1FSEjI/TCGhggJCcGhQ4c0DtOrVy+kpqaqiueFCxewfft2PPfcczVOp7S0FIWFhWoPIiKp6e2GeteuXYNSqYSDg4Nau4ODAzIyMjQOM3r0aFy7dg19+vSBEAJ3797Fa6+9VutqfmxsLObOnStpdiKiBzWpu5MmJyfj448/xldffYXAwEBkZmYiMjISH374Id5//32Nw0RHRyMqKkr1urCwEC4uLo0VmZqpnJwc5OTk1Lu/k5MTnJx4R9XmTG/F1M7ODkZGRsjLy1Nrz8vLg6Ojo8Zh3n//fbz88st49dVXAQBeXl4oKSnBhAkT8N5778HQsPpWC4VCAYVCIf0HoEfawoULsXjx4nr3j4qKwqJFi3SYiPRNb9tMTUxM4Ofnh6SkJFVbRUUFkpKSEBQUpHGYW7duVSuYRkZGAAAhhO7CEv3PnDn3HgcPajfcwYP3hqPmS6+r+VFRUQgLC4O/vz8CAgIQFxeHkpIShIeHAwDGjh2LNm3aIDY2FgAwcOBALF68GN27d1et5r///vsYOHCgqqgSNYZevd5Ct25j6t3f0pKr+M2dXovpiBEjcPXqVcyePRu5ubnw8fFBYmKiaqdUVlaW2pLorFmzYGBggFmzZuHy5cto3bo1Bg4ciI8++khfH4EeUVZWTrCyYoGk+/S+A2rKlCmYMmWKxveSk5PVXrdo0QIxMTGIiYlphGRERPWn92Iqd0VFOSgurv9eW0tLLrEQPYpYTOuQmvoNfv21/sepBgfHoH//OboLRESyxGJaBz+/iXB3H6R6XV5+GwkJfQAA4eEHYGxsptafOxqIHk0spjXonzxHY/ttZRkS/vf8hXPbYGZkUsMYNA9PRM2T3i/BR0TUHHDJtA75pUXILytWvS6tKFc9zyzOhcLQWK2/rYklbBVWjZaPiOSBxbQOW3NSserPXzW+Ny09oVpbmGswxrn113EqIpIbFtM6DHTyQy9b93r3tzWx1GEaIpIrFtM62CqsZLHazuNdieSNxbSJ4PGuRPLGYtpE8HhXInljMW0iHrywRllZieq5o6MPTEws9BGLiP6HxVTGajpxALh38kDs/5733f8RTx4g0jMW0yaCx7sSyRuLaRPB412J5I3FtIng8a5E8sZi2kTI5XhXItKMFzohIpIAiykRkQS4mk9EzUJ6ejpOnjxZ7/6enp7w8fGRbPospkTUZM2Zc//5ypXT8WcNR7xo4uoajHHjktXG8TBYTImoWQgNjcO1a/VfMrWz85R0+iymRNQsODn5wMnJR2/T5w4oIiIJsJgSEUmAxZSISAIspkREEmAxJSKSAIspEZEEWEyJiCTAYkpEJAEWUyIiCbCYEhFJgMWUiEgCLKZERBLQ+kInbm5ueOWVVzBu3Di0a9dOF5lIpoqKclBcnFPv/paWTrCyctJhIiL50LqYTp8+HStXrsQHH3yAJ554AuPHj8fQoUOhUCgaFCA+Ph4LFixAbm4uvL298cUXXyAgIKDG/jdu3MB7772H//znPygoKICrqyvi4uLw3HPPNWj6VH+pqd/g11/n1rt/cHAM+vefo7tAhJycHOTk1P8/OCcnJzg58T84XWhQMZ0+fTrS0tKwcuVKTJ06FZMmTcLo0aPxyiuvwNfXt97jWr9+PaKiorBkyRIEBgYiLi4OoaGhOHPmDOzt7av1Lysrw9NPPw17e3ts3LgRbdq0wZ9//onHHntM249BDeDnNxHu7oNUr8vLbyMhoQ8AIDz8AIyNzdT6W1ryj1bXFi5ciMWLF9e7f1RUFBYtWqTDRI+uBl/P1NfXF76+vli0aBG++uorzJgxA19//TW8vLwwbdo0hIeHw8DAoNZxLF68GBEREQgPDwcALFmyBNu2bcOKFSswc+bMav1XrFiBgoICHDx4EMbGxgDubXYg3eqfPEdj+21lGRL+9/yFc9tgZmRSwxg0D08NU/XK8AcPajfswYP3hpfq6vJ0X4OLaXl5OTZv3oyEhATs2rULPXv2xPjx4/HXX3/h3Xffxe7du7FmzZoahy8rK0Nqaiqio6NVbYaGhggJCcGhQ4c0DvPTTz8hKCgIkydPxo8//ojWrVtj9OjRmDFjBoyMjDQOU1paitLSUtXrwsLCBn5iIvnp1estdOs2pt79ubagO1oX07S0NCQkJGDt2rUwNDTE2LFj8emnn6Jz586qPkOHDkWPHj1qHc+1a9egVCrh4OCg1u7g4ICMjAyNw1y4cAF79uzBmDFjsH37dmRmZmLSpEkoLy9HTEyMxmFiY2Mxd279t/NRzfJLi5BfVqx6XVpRrnqeWZwLhaGxWn9bE0vYKqwaLd+jyMqKO/kq6XsHqdbFtEePHnj66afx9ddfY8iQIarV7arat2+PkSNHShKwqoqKCtjb2+Pbb7+FkZER/Pz8cPnyZSxYsKDGYhodHY2oqCjV68LCQri4uEie7VGwNScVq2q4Ydm09IRqbWGuwRjn1l/HqYjuOXhwIX77rf7bj3v2jEJoqHTbj7UuphcuXICrq2utfSwsLJCQUP2Pqyo7OzsYGRkhLy9PrT0vLw+Ojo4ah3FycoKxsbHaKr2Hhwdyc3NRVlYGE5Pq2+wUCkWDjzQgdQOd/NDL1r3e/W1NLHWYhkh9e/6p7IP4TYthXbIP/m/4OXX0rB+ti+mVK1eQm5uLwMBAtfbDhw/DyMgI/v7+9RqPiYkJ/Pz8kJSUhCFDhgC4t+SZlJSEKVOmaBymd+/eWLNmDSoqKmBoeO98g7Nnz8LJyUljISVp2SqsuNpOsjXCpRdCHLrVu7/U/9lrfQbU5MmTkZ2dXa398uXLmDx5slbjioqKwtKlS7Fq1SqcPn0ar7/+OkpKSlR798eOHau2g+r1119HQUEBIiMjcfbsWWzbtg0ff/yx1tMloubHVmGFf1g51fsh9YKB1kump06d0ngsaffu3XHq1CmtxjVixAhcvXoVs2fPRm5uLnx8fJCYmKjaKZWVlaVaAgUAFxcX7NixA2+88Qa6deuGNm3aIDIyEjNmzND2YxARSUrrYqpQKJCXl4cOHTqotefk5KBFC+2PtJoyZUqNq/XJycnV2oKCgvDbb9psGSEi0j2tV/OfeeYZREdH4+bNm6q2Gzdu4N1338XTTz8taTgioqZC60XJhQsXol+/fnB1dUX37t0BAOnp6XBwcMB3330neUAioqZA62Lapk0bHD9+HKtXr8axY8dgZmaG8PBwjBo1SuMxp0REj4IGnU5qYWGBCRMmSJ2FiKjJavC5+adOnUJWVhbKysrU2gcNGlTDEEREzVeDzoAaOnQo/vjjDxgYGEAIAQCqK0QplUppExIRNQFa782PjIxE+/btceXKFZibm+PkyZPYt28f/P39NR7KRET0KNB6yfTQoUPYs2cP7OzsYGhoCENDQ/Tp0wexsbGYNm0ajh49qoucRESypnUxVSqVsLK6dxqWnZ0d/v77b7i7u8PV1RVnzpyRPCARyVt6ejpOnjxZ7/6enp7w8fHRXSA90bqYdu3aFceOHUP79u0RGBiITz75BCYmJvj222+rnRVFRM1X5dX6V66cjj9ruDSjJq6uwRg3LlltHM2B1sV01qxZKCkpAQB88MEH+Ne//oW+ffvC1tYW69evlzwgEclbaGgcrl2r/5KpnZ2nDtPoj9bFNDQ0VPW8Y8eOyMjIQEFBAVq2bFnnPZ+IqPlxcvKBk5OPvmPonVZ788vLy9GiRQucOHFCrb1Vq1YspET0SNOqmBobG6Ndu3Y8lpSI6AFaH2f63nvv4d1330VBQYEu8hARNUlabzP98ssvkZmZCWdnZ7i6usLCwkLt/bS0NMnCERE1FVoX08r7NRER0X1aF9OabqlMRPQo03qbKRERVaf1kqmhoWGth0FxTz/pWlFRDoqLc+rd39LSCVZWTjpMRNSAYrp582a11+Xl5Th69ChWrVqFuXPnShaMqCapqd/g11/r/1sLDo5B//5zdBeICA0opoMHD67W9vzzz8PT0xPr16/H+PHjJQlGVBM/v4lwd79/EfLy8ttISOgDAAgPPwBjYzO1/paWXCol3WvwlfYf1LNnT97KhBqFlZX6antZWYnquaOjD0xMLDQNRqRTkhTT27dv4/PPP0ebNm2kGB1RNf2T59T43m1lGWL/97zv/o9gZmRSQ8+ax0H0sLQupg9e0EQIgaKiIpibm+P777+XNBwRUVOhdTH99NNP1YqpoaEhWrdujcDAQLRs2VLScESa5JcWIb+sWPW6tKJc9TyzOBcKQ/VbjtuaWMJWYdVo+ejRpHUxHTdunA5iENXf1pxUrKrhYsTT0hOqtYW5BmOcW38dp6JHndbFNCEhAZaWlnjhhRfU2n/44QfcunULYWFhkoUj0mSgkx962brXu7+tiaUO0xDdo3UxjY2NxTfffFOt3d7eHhMmTGAxJZ2zVVhxtZ1kR+vTSbOystC+fftq7a6ursjKypIkFBFRU6N1MbW3t8fx48ertR87dgy2traShCIiamq0LqajRo3CtGnTsHfvXiiVSiiVSuzZsweRkZEYOXKkLjISEcme1ttMP/zwQ1y6dAlPPfUUWrS4N3hFRQXGjh2Ljz/+WPKARERNgdbF1MTEBOvXr8e///1vpKenw8zMDF5eXnB1ddVFPiKiJqHBp5N26tQJnTp1kjILEVGTpfU20+HDh2P+/PnV2j/55JNqx54SET0qtC6m+/btw3PPPVet/dlnn8W+ffsaFCI+Ph5ubm4wNTVFYGAgjhw5Uq/h1q1bBwMDA96Xioj0TutiWlxcDBOT6lflMTY2RmFhodYB1q9fj6ioKMTExCAtLQ3e3t4IDQ3FlStXah3u0qVLeOutt9C3b1+tp0lEJDWti6mXlxfWr19frX3dunXo0qWL1gEWL16MiIgIhIeHo0uXLliyZAnMzc2xYsWKGodRKpUYM2YM5s6diw4dOmg9TSIiqWm9A+r999/HsGHDcP78eTz55JMAgKSkJKxZswYbN27UalxlZWVITU1FdHS0qs3Q0BAhISE4dOhQjcN98MEHsLe3x/jx47F///5ap1FaWorS0lLV64YsPRMR1UXrYjpw4EBs2bIFH3/8MTZu3AgzMzN4e3tjz549aNWqlVbjunbtGpRKJRwcHNTaHRwckJGRoXGYAwcOYPny5UhPT6/XNGJjY3lvKiLSuQbd6nnAgAFISUlBSUkJLly4gBdffBFvvfUWvL29pc6npqioCC+//DKWLl0KOzu7eg0THR2Nmzdvqh7Z2dk6zUhEj6YGH2e6b98+LF++HJs2bYKzszOGDRuG+Ph4rcZhZ2cHIyMj5OXlqbXn5eXB0dGxWv/z58/j0qVLGDhwoKqtoqICANCiRQucOXMGjz/+uNowCoUCCoVCq1xERNrSqpjm5uZi5cqVWL58OQoLC/Hiiy+itLQUW7ZsadDOJxMTE/j5+SEpKUl1eFNFRQWSkpIwZcqUav07d+6MP/74Q61t1qxZKCoqwmeffQYXFxetMxARSaHexXTgwIHYt28fBgwYgLi4OPzzn/+EkZERlixZ8lABoqKiEBYWBn9/fwQEBCAuLg4lJSUIDw8HAIwdOxZt2rRBbGwsTE1N0bVrV7XhH3vsMQCo1k5E1JjqXUx/+eUXTJs2Da+//rqkp5GOGDECV69exezZs5GbmwsfHx8kJiaqdkplZWXB0LBBm3aJiBpNvYtp5V50Pz8/eHh44OWXX5bskntTpkzRuFoPAMnJybUOu3LlSkkyEBE9jHov8vXs2RNLly5FTk4OJk6ciHXr1sHZ2RkVFRXYtWsXioqKdJmTiEjWtF5/trCwwCuvvIIDBw7gjz/+wJtvvol58+bB3t4egwYN0kVGIiLZa/ChUQDg7u6OTz75BLGxsdi6dWutp4ASNSdFRTkoLs6pd39LSydYWTnpMBHp20MV00pGRkYYMmQIr95Ej4zU1G/w66/1P7MuODgG/fvP0V0g0jtJiinRo8bPbyLc3e9v1iovv42EhD4AgPDwAzA2NlPrb2nJpdLmjsWUqAGsrNRX28vKSlTPHR19YGJioY9YpEcspkRa6J88R2P7bWUZYv/3vO/+j2BmVP2av/doHp6aPhZTogbILy1Cflmx6nVpRbnqeWZxLhSGxmr9bU0sYauwarR81PhYTIkaYGtOKlb9+avG96alJ1RrC3MNxji3/jpORfrEYkrUAAOd/NDL1r3e/W1NLHWYhuSAxZSoAWwVVlxtJzW8gggRkQRYTImIJMBiSkQkARZTIiIJsJgSEUmAxZSISAIspkREEmAxJSKSAA/aJ6KHwgtl38NiStSEyaGQ8ULZ97CYEjVh+ixklZcjbFuUgz6dh6nayyrKseDsVgDA2/8YCJMHrqDlVpSDjqpLGUqTRQ5YTImaMDlc8f9A/pkar6BVWVSrCnMNRkeu5hORvtV0gWrg3kWqKy8A+MK5bY1ykWpeQeseFlOiJkwOF6nmFbTuYTElasJ4kWr5YDElasK4ii0fLKZETRhXseWDZ0AREUmAxZSISAIspkREEmAxJSKSAIspEZEEWEyJiCTAYkpEJAEWUyIiCciimMbHx8PNzQ2mpqYIDAzEkSNHauy7dOlS9O3bFy1btkTLli0REhJSa38iosag92K6fv16REVFISYmBmlpafD29kZoaCiuXLmisX9ycjJGjRqFvXv34tChQ3BxccEzzzyDy5cvN3JyIqL79F5MFy9ejIiICISHh6NLly5YsmQJzM3NsWLFCo39V69ejUmTJsHHxwedO3fGsmXLUFFRgaSkpEZOTkR0n16LaVlZGVJTUxESEqJqMzQ0REhICA4dOlSvcdy6dQvl5eVo1aqVxvdLS0tRWFio9iAikppei+m1a9egVCrh4OCg1u7g4IDc3Nx6jWPGjBlwdnZWK8hVxcbGwsbGRvVwcXF56NxERA/S+2r+w5g3bx7WrVuHzZs3w9TUVGOf6Oho3Lx5U/XIzs5u5JRE9CjQ6yX47OzsYGRkhLy8PLX2vLw8ODo61jrswoULMW/ePOzevRvdunWrsZ9CoYBCoZAkLxFRTfS6ZGpiYgI/Pz+1nUeVO5OCgoJqHO6TTz7Bhx9+iMTERPj7+zdGVCKiWun94tBRUVEICwuDv78/AgICEBcXh5KSEoSHhwMAxo4dizZt2iA2NhYAMH/+fMyePRtr1qyBm5ubatuqpaUlLC15FXEi0g+9F9MRI0bg6tWrmD17NnJzc+Hj44PExETVTqmsrCwYGt5fgP76669RVlaG559/Xm08MTExmDNnTmNGJyJS0XsxBYApU6ZgypQpGt9LTk5We33p0iXdByIi0lKT3ptPRCQXLKZERBJgMSUikgCLKRGRBFhMiYgkwGJKRCQBFlMiIgmwmBIRSYDFlIhIAiymREQSYDElIpIAiykRkQRYTImIJMBiSkQkARZTIiIJsJgSEUmAxZSISAIspkREEmAxJSKSAIspEZEEWEyJiCTAYkpEJAEWUyIiCbCYEhFJgMWUiEgCLKZERBJgMSUikgCLKRGRBFhMiYgkwGJKRCQBFlMiIgmwmBIRSYDFlIhIAiymREQSYDElIpIAiykRkQRYTImIJCCLYhofHw83NzeYmpoiMDAQR44cqbX/Dz/8gM6dO8PU1BReXl7Yvn17IyUlItJM78V0/fr1iIqKQkxMDNLS0uDt7Y3Q0FBcuXJFY/+DBw9i1KhRGD9+PI4ePYohQ4ZgyJAhOHHiRCMnJyK6T+/FdPHixYiIiEB4eDi6dOmCJUuWwNzcHCtWrNDY/7PPPsM///lPvP322/Dw8MCHH34IX19ffPnll42cnIjovhb6nHhZWRlSU1MRHR2tajM0NERISAgOHTqkcZhDhw4hKipKrS00NBRbtmzR2L+0tBSlpaWq1zdv3gQAFBYW1pqt5G5pre/Xpa7x18fDZpBLDjlkkEsOOWRoTjl0naHyPSFE3SMSenT58mUBQBw8eFCt/e233xYBAQEahzE2NhZr1qxRa4uPjxf29vYa+8fExAgAfPDBBx8NfmRnZ9dZz/S6ZNoYoqOj1ZZkKyoqUFBQAFtbWxgYGDRonIWFhXBxcUF2djasra2litokc8ghg1xyyCGDXHLIIYMUOYQQKCoqgrOzc5199VpM7ezsYGRkhLy8PLX2vLw8ODo6ahzG0dFRq/4KhQIKhUKt7bHHHmt46Cqsra31+kORUw45ZJBLDjlkkEsOOWR42Bw2Njb16qfXHVAmJibw8/NDUlKSqq2iogJJSUkICgrSOExQUJBafwDYtWtXjf2JiBqD3lfzo6KiEBYWBn9/fwQEBCAuLg4lJSUIDw8HAIwdOxZt2rRBbGwsACAyMhLBwcFYtGgRBgwYgHXr1uH333/Ht99+q8+PQUSPOL0X0xEjRuDq1auYPXs2cnNz4ePjg8TERDg4OAAAsrKyYGh4fwG6V69eWLNmDWbNmoV3330XnTp1wpYtW9C1a9dGy6xQKBATE1Nt80Fjk0MOOWSQSw45ZJBLDjlkaOwcBkLUZ58/ERHVRu8H7RMRNQcspkREEmAxJSKSAIspEZEEWEyJiCSg90OjmoITJ07g5MmTaNWqFTw8PNC2bdtHMoOccsgpixxyyCGDHPM0ao76XJDkUZWeni6effZZYWpqKnx9fYWZmZno2bOn2L179yOVQU455JRFDjnkkEGOefSRg8W0Bu+9955o0aKF6Nevn9iyZYvIyMgQmZmZ4v333xft27d/ZDLIKYecssghhxwyyDGPvnKwmD7g9u3bYtKkScLZ2Vl8/fXXoqioSO39zMxM4erqKg4dOtSsM8gph5yyyCGHHDLIMY++c7CYPiAtLU04OTmJZcuWidLS0mrv//DDD6Jt27YiJyenWWeQUw45ZZFDDjlkkGMefedgMX3AsmXLRGBgoCgpKVFrv3v3rti4caPo0KGDiIiIaPYZ5JRDTlnkkEMOGeSYR985uDf/AQqFArm5uVAqlaq29PR0bN++HTt27ICPjw/eeuutZp9BTjnklEUOOeSQQY559J2DFzrRwNPTEy4uLvD390dWVhauXLmC7Oxs9O3bF9HR0XB1dX0kMsgph5yyyCGHHDLIMY8+c7CYanD69Gls2LAB33//PTw9PdG+fXsMGTIEwcHBAO7dyqChtzxpShnklENOWeSQQw4Z5JhHmxxSZ2Ix1VJFRQUMDQ2hVCphZGT0yGaQUw45ZZFDDjlk0JRH33SdQ/+fUOb27NmDy5cvA7j3ZYh7O+1UP9Jbt27V7zawTTyDnHLIKYsccsghQ215Hixgd+7cQUZGBs6ePavXHNevX8fGjRuRmJgozQR1tmurGfj555+FnZ2d+Pjjj4UQ9/YKVjp48KAYNWqUGDJkiOjRo4eIi4urdlxbc8kgpxxyyiKHHHLIUFseTfLy8sTUqVNFSEiIuH79eqPnKC4uFikpKWLatGnCxsZGGBgYiCtXrjz0NLlkWosBAwZgypQpCAkJAQAYGRnh6tWrGD58OIKDg3Hz5k20b98evr6+WLduHV555ZVmmUFOOeSURQ455JChtjwPunv3Luzt7TFp0iQUFxdj8uTJOssxefJkPPnkkwCgWjJXKpU4dOgQ3nzzTaxevRpr167F0KFDpZkvD12OHyHbt28X//jHP4Sfn59YvXq1yMrKUr137NgxYWVlJU6cONHsM8gph5yyyCGHHDJoUnWJWQghFi5cKBwdHYW3t7ckS4XaOHv2rBg5cqR49tlnhRBC/Pnnn8LU1FQcOHDgocbLYlpPt27dEoMHDxYvvPCC+P3336v9OC5cuCC8vLzEkSNHmnUGOeWQUxY55JBDhqru3LkjKioqREVFhartp59+Ej169BAdO3YUM2fOFHv37hVlZWU6y3Dp0iXx9ttvi8OHD6u1Z2dni44dO4off/xRCCHElClTRLt27R5qWiym9ZSUlCTMzc1FSkpKtfdu3LghIiIihKOjo/j777+bdQY55ZBTFjnkkEOGSgUFBSIoKEikp6cLIYQ4efKkGDJkiGjbtq0YM2aM2Lp1qygoKNB5jvPnz4suXbqI/fv3CyGEUCqVQqlUiuLiYvH000+Ljz76SAghRFlZ2UOfs88zoOrpzz//RLdu3dCzZ09VW0VFBTIyMrBhwwYcPnwY8+fPh5OTU7POIKcccsoihxxyyFCpZcuWaNGiBaZPnw4/Pz+sX78e7u7umDt3LkJCQtCuXTtVX6HDY1A7dOgADw8PLF68GH369FHt0TcwMMDhw4cxatQoAICxsbHafGuQh638j4q8vDxhZmYmVq1aJa5fvy5+//13sWTJEtGvXz9hb28vPvroo2qrK1VXb5pLBjnlqC1LcHCwsLe3Fx9//LEoLy9X9X/wvG195Lh79664evVqo2f48MMP1b6X7OxsyTNUdeXKFWFiYiKcnZ3FJ598Io4fPy6USqXqfV39Jh6Um5srHB0dxUcffSQuXLggdu3aJYYNGya8vb1Fbm6uZNNhMdXCwoULhb+/vzA2NhZdu3YV7u7uYtiwYeLs2bOqPvv37xc7duwQWVlZOtkWJIcMcspRU5ahQ4eKjIwMVZ87d+6IzZs3i4ULF+rscJz65Lh27ZoYPHiwGDlypE52vNSU4fTp06o+J06cEO+++64ICAjQ+Sr/Bx98IGxtbcXly5dVbY1VRKtatWqVeOqpp4SBgYFwc3MTISEh4rvvvpM0D8+A0tKlS5dw9OhRtGjRAq1bt1atGixbtgwLFy7E5cuX8dhjj8HCwgIDBw7EggULmmUGOeWomsXY2BiOjo7w9/cHcP8gdiMjI2zduhWff/45XF1dsWzZMp3lSEtLg4mJSY05tm3bhpUrV0KpVOI///lPo2W4efMm1q5diy1btuDu3bs4evQounfvjt27d0ueoSp7e3vMmTMHkyZN0ul06lJUVIT9+/fDyMgIbdu2haenp7QTkKQkP8IyMzNF3759hbGxsXjttddEWlqaSE1NFYmJicLS0lIkJSU9EhnklKNS5d7sqquWX3/9tTAwMGi0CycLITReW/PcuXPC1tZWtWOkMZw+fVpMmDBBtGzZUhw4cEDk5+cLCwsLsWPHDp1Od/fu3WLr1q06nYYcsJg2QOUf5969e0WrVq1Enz59NB67N2LECBEeHt5sM8gpR1WaVtsKCgrEO++8IywsLERAQIBIS0vTaYbS0lK17XGVmaoW9qFDh4r33ntPZxmUSqU4c+aMWtvx48fF448/LjZv3iyEuHeLj4c9JIju4RlQDVC5R3DVqlXo1asXNm/eDE9PTyiVSrVrKV64cAHdunVrthnklAO4t1d43759uHTpklrb8uXL4enpidWrV+PTTz/Fnj170L17d53luHnzJjw9PREQEICwsDAsW7YMGRkZaueH5+fn4/jx4+jYsaNOMpSWlmLSpElYsmQJ7t69q2r38vLCa6+9hi+++AIA8O9//xuPP/44UlJSdJLjUcJi2kCXL19GWloa/vWvf8HOzg4VFRUwMjKCkZER8vPzMWnSJBw/fhyBgYF6z9CjRw+dZdAmhy7nBXDvcJdt27ZhxIgRqsIaGBiIKVOm4KWXXsLvv/+OiIgIWFhY6DSHjY0N+vXrh+zsbLi5ueHTTz9FSEgIfH19MXz4cLzxxht44oknUFxcDF9fX51kUCgUMDMzw+nTp9Gixb0jICv/cyssLIQQAkVFRQCATZs2oXfv3jrJ8UjR74Jx03Xr1i3h4uKiOoNCiHurcv/5z3/E888/LwIDA8XKlSubfQY55RBCiOvXrwuFQiF8fX2FsbGxGDp0qDh27FijTLuqmzdvihYtWoiDBw8KIe7tQV+zZo2YMmWKeO6558SkSZN0fmjSzZs3hbm5udi8ebPa0RR9+vQRgwYN0um0H0Xcm/8Q/v3vf2Pp0qUIDw+Ho6Mjtm3bhqtXr+Kxxx7DiBEjMHr0aCgUCgDA2bNn4ezsDEtLy0bNMGrUKJiamiIlJQUFBQXo27cvHnvsMUkz1JXj+eefx+jRo2Fubg4A2Lp1K/z8/ODs7Cx5DgCIj4/H1KlTsWbNGrzwwgt6u6ZnTEwM1q5dW+1Sc1WvMyp0fNHkzz77DCtWrEC3bt3g5eWFZcuWQalU4scff0TXrl11Nt1Hkp6LeZMXGxsr+vTpI5ycnMTo0aPFggULxLlz59T6/Prrr+LFF18Us2fP1luG7777TvTt21e8+uqrOslQNYejo6MYM2aMWLhwoVqO1NRUMWjQIGFgYCCefPJJneUQQgh7e3sxc+ZMnU6jPmxsbERsbKwQQn3nU9XnurZ69Wrx5JNPChcXFzFx4kS1UytJOiymEiguLhZCqJ9dU1FRoXbmzZIlS4STk5P45ZdfdJKhsLBQLUulysODysvLxeHDh4W9vb3YsmWLTjIIIVTXy6w6L8rKykR8fLywtLQUgwYNEt98840wMzMTmzZt0lmOLVu2iGeffbZRzv+uK8fMmTP1Xrju3Lmj9i9Jj8VUAg9enadqERXi3lkvCQkJwtjYWPj7+zdKpppO55w3b57o3bu3zqdfWdwri8jGjRuFt7e36lCguLg44eTkpNMMp06d0un460tXp7BqQx9nHT1quDdfApXbv0pKSgBAtfe0sLAQu3fvxtSpUxEZGYmBAwdiw4YNOslQUVGBnTt3qg7DuXPnjtp7ldvlFAoFTExMkJ+fr5McAHDx4kVs3boVN27cUB0KNHz4cEyfPh2rV69GRUUFIiMjYWhoiM8//1xnOTw8PHQ2bm1UbivWp8a8ud6jileNksjevXuxYcMGfPnll1AqlTh58iTWrVuH//u//4O9vT1++OEHPPPMMwB0c2MvQ0NDpKenY+bMmfD09MStW7fg7++P559/Ht26dYOlpSVycnKwfft2tGvXDra2tpJOv6rDhw9j0aJF6NOnj9rOruDgYNjb22PHjh149tln8fPPP+PcuXM6y0HUmLg3XyLl5eWwtrbGO++8A0dHR8THxyM/Px/R0dGYNm2aqp8u7xhZXl4OBwcHREdHo0uXLvjxxx+RmpqKgoIC3Lx5E25ubiguLsaSJUvw5JNP6mxPcnl5OVxcXPDFF1/ghRdeULXv3LkTI0eOxN69e+Ht7S35dIn0Ss+bGZqVL774QhgYGAgbGxsxadIktW1lD25X1ZX58+eLTp06qbWdPHlS7Nu3T2zZskXcunWrUXLExcWJTp06iX379on8/Hzx559/ikGDBglfX99qlz3j9jxqDrhkKjE/Pz94eHjg+++/B3DvBmKV21AbS8eOHREWFob3339f4/uNdR/ziIgI7Nu3DwYGBsjOzkabNm2wYcMG+Pj46HzaRI2NxVRildsDjx07hq5du+plw/8vv/yCAQMG4OrVqzrdNlqX4uJinDp1Cj///DM6deqEYcOGwcLCotGKOVFjYjHVgUWLFmHAgAHo3Lmz3jJs2LABL774ot6mD+j+7B4iOWEx1QEWkeo4T6i5YzElIpIAN1wREUmAxZSISAIspkREEmAxJSKSAIspEZEEWEyJiCTAYkpEJAEW0waYM2eOJOeXr1y5Uif3Y6qP5ORkGBgY4MaNGw81nv79+2P69Om19nFzc0NcXJzqtYGBAbZs2QIAuHTpEgwMDJCenv5QORpqzpw5cHBwUMv0oJSUFHh5ecHY2BhDhgyp97wbN24chgwZInnmBwkhMGHCBLRq1Uqv87Kh9Pl3UF8P/s5v3bqF4cOHw9ra+v5vQR9XV2nqioqKxLVr1x56PLdu3RJ5eXkSJNLe3r17BQBx/fr1hxpPcHCwiIyMrLXPlStX1K6gBUBs3rxZCHHvalo5OTmquxNIlas+Tp06pcqSk5NT4y09AgICxEsvvSSys7PF9evXRWlpqcjJyanzalc3btxolM+xfft2YWxsLFJSUtTmZVORkJAgbGxs9B1DCFHz7y8/P1919wghhPjqq69E69atxR9//KH6LfDi0FoQQkCpVMLS0vKh7zJaXl4OMzMzmJmZSZTunsqMjX2lqtq0bt26xveMjIzg6OjYiGnuO3/+PABg8ODBtZ7qev78ebz22mto27atqq22zEqlEgYGBrCxsZEubC3Onz8PJycn9OrVq8HjkOPvRk5atWql9vr8+fPw8PBQv8NrY1b9xvTLL7+I3r17CxsbG9GqVSsxYMAAkZmZqXr/4sWLAoBYu3atCAoKEgqFQnh6eork5GRVn8r/pbZv3666D/vevXtFTEyM8Pb2VvVTKpVi7ty5ok2bNsLExER4e3ur3Tivclrr1q0T/fr1EwqFQiQkJNT5P/LDZLxz546YOnWqaN26tVAoFKJ3797iyJEj1Yb7+eefhZeXl1AoFCIwMFD88ccfqj7Xrl0TI0eOFM7OzsLMzEx07dpVrFmzRi1jcHCwmDx5spg8ebKwtrYWtra2YtasWWpLba6uruLTTz9VvUaVJdPKz3j06FHV86qPsLAwsWrVKtGqVatqS46DBw8WL730Uo3z7/jx4+KJJ54QpqamolWrViIiIkJ1w7+YmJhq06pp/ld9JCQkVFt6qfwef/zxR+Hh4SGMjIzExYsXRVhYmBg8eLDavJo6dap4++23RcuWLYWDg4OIiYlRm+bp06dF7969hUKhEB4eHmLXrl1q8+tBYWFhavlcXV2FEKLe3/+DvxtNsrKyxAsvvCBsbGxEy5YtxaBBg8TFixdV7x85ckSEhIQIW1tbYW1tLfr16ydSU1PVxnH9+nUxYcIEYW9vr/odb926VW3+JSYmis6dOwsLCwsRGhoq/v77b415quZPTEwUPj4+wtTUVDzxxBMiLy9PbN++XXTu3FlYWVmJUaNGqa0V1TZfavr9VX53lWtgwcHBan2Cg4OFEM34hnobN24UmzZtEufOnRNHjx4VAwcOFF5eXqobvFXOuLZt24qNGzeKU6dOiVdffVVYWVmpVuErv7Bu3bqJnTt3iszMTJGfn1+tmC5evFhYW1uLtWvXioyMDPHOO+8IY2NjcfbsWbVpubm5iU2bNokLFy6Iv//+u97FtCEZp02bJpydncX27dvFyZMnRVhYmGjZsqXIz89XG87Dw0Ps3LlTHD9+XPzrX/8Sbm5uqpvx/fXXX2LBggXi6NGj4vz58+Lzzz8XRkZG4vDhw6qMwcHBwtLSUkRGRoqMjAzx/fffC3Nzc/Htt9+q+tS3mN69e1ds2rRJABBnzpwROTk54saNG+LWrVvCxsZGbNiwQTWOvLw80aJFC7Fnzx6N8664uFg4OTmJYcOGiT/++EMkJSWJ9u3bq/44ioqKREJCggAgcnJyRE5OTrVxVG6CsLa2FnFxcSInJ0fcunVLYzE1NjYWvXr1EikpKSIjI0OUlJRoLKbW1tZizpw54uzZs2LVqlXCwMBA7Ny5UzU9d3d38fTTT4v09HSxf/9+ERAQUGsxvXHjhvjggw9E27ZtRU5Ojrhy5YoQQtT7+3/wd/OgsrIy4eHhIV555RVx/PhxcerUKTF69Gjh7u4uSktLhRBCJCUlie+++06cPn1anDp1SowfP144ODio3VSxZ8+ewtPTU+zcuVOcP39ebN26VWzfvl1t/oWEhIj//ve/IjU1VXh4eIjRo0dr/MxV8/fs2VMcOHBApKWliY4dO4rg4GDxzDPPiLS0NLFv3z5ha2sr5s2bpxqutvlS0++v8rurLKb5+fkiIiJCBAUFiZycHNV8a7bF9EFXr14VAFRLXpV/xFVndHl5uWjbtq2YP3++EOL+F/bgrZEfLKbOzs7io48+UuvTo0cPMWnSJLVpxcXFqfWpbzHVNmNxcbEwNjYWq1evVrWVlZUJZ2dn8cknn6gNt27dOlWf/Px8YWZmJtavX19jpgEDBog333xT9To4OFh4eHioLYnOmDFDeHh4qF7Xt5hWzfXgNqvXX39dPPvss6rXixYtEh06dKhxu+W3334rWrZsqXbr623btglDQ0PVlf43b96scYn0QTY2NiIhIUH1WlMxBSDS09PVhtNUTPv06aPWp0ePHmLGjBlCiHtrUy1atFAr7HUtmQohxKeffqpaIhVCu++/rtt+f/fdd8Ld3V1tPpeWlgozMzOxY8cOjcMolUphZWWlWvLcsWOHMDQ0FGfOnNHYv3L+VV1zjI+PFw4ODjXmqsy/e/duVVtsbKwAIM6fP69qmzhxoggNDRVCaDdfHvz9PbhvIDIyUrVEWqnZ7s0/d+4cRo0ahQ4dOsDa2hpubm4AgKysLLV+QUFBquctWrSAv78/Tp8+rdbH39+/xukUFhbi77//Ru/evdXae/furdV4aqNtxvPnz6O8vFwtk7GxMQICAqoNV3XcrVq1gru7u6qPUqnEhx9+CC8vL7Rq1QqWlpbYsWNHtXnYs2dPtW2OQUFBOHfuHJRKZYM+ryYRERHYuXMnLl++DODeHuBx48bVuK3z9OnT8Pb2hoWFhaqtd+/eqKiowJkzZyTLVcnExATdunWrs9+DfZycnHDlyhUAwJkzZ+Di4qK2PTYgIEDrLNp8/3X9Jo8dO4bMzExYWVmp9hW0atUKd+7cUW1zzsvLQ0REBDp16gQbGxtYW1ujuLhY9TtJT09H27Zt8Y9//KPG6Zibm+Pxxx9Xva46X2pTdX46ODjA3NwcHTp0UGurHI8286Uhmu3W5oEDB8LV1RVLly6Fs7MzKioq0LVrV5SVlWk9rqp/kA9DqvE01rgXLFiAzz77DHFxcfDy8oKFhQWmT5/eoHn4sLp37w5vb2/83//9H5555hmcPHkS27Zta/QcNTEzM6vX9VqNjY3VXhsYGKCiokJXsepU1++muLgYfn5+WL16dbX3KncshoWFIT8/H5999hlcXV2hUCgQFBSk+p3UZyerpvki6nF10KrDGRgY6HX+Nssl0/z8fJw5cwazZs3CU089BQ8PD1y/fl1j399++031/O7du0hNTdXqfuvW1tZwdnZGSkqKWntKSgq6dOnSsA/wkBkff/xxmJiYqGUqLy/Hf//732qZqo77+vXrOHv2rGrcKSkpGDx4MF566SV4e3ujQ4cOOHv2bLXpHT58uNo4O3Xq1KC7sJqYmACAxqXaV199FStXrkRCQgJCQkLg4uJS43g8PDxw7NgxlJSUqNpSUlJgaGgId3d3rXM1Bnd3d2RnZyMvL0/V9t///lfr8Wjz/dfF19cX586dg729PTp27Kj2qDxaISUlBdOmTcNzzz0HT09PKBQKXLt2TTWObt264a+//tL422lM9Zkvtf3+6tIsi2nLli1ha2uLb7/9FpmZmdizZw+ioqI09o2Pj8fmzZuRkZGByZMn4/r163jllVe0mt7bb7+N+fPnY/369Thz5gxmzpyJ9PR0REZGSvFxtM5oYWGB119/HW+//TYSExNx6tQpRERE4NatWxg/frxa3w8++ABJSUk4ceIExo0bBzs7O9WB5p06dcKuXbtw8OBBnD59GhMnTlT7Q6+UlZWFqKgonDlzBmvXrsUXX3zR4M/u6uoKAwMD/Pzzz7h69SqKi4tV740ePRp//fUXli5dWud3NGbMGJiamiIsLAwnTpzA3r17MXXqVLz88stwcHBoUDZde/rpp/H4448jLCwMx48fR0pKCmbNmgUAWt2lQJvvvy5jxoyBnZ0dBg8ejP379+PixYtITk7GtGnT8NdffwG49zv57rvvcPr0aRw+fBhjxoxRWxoNDg5Gv379MHz4cOzatQsXL17EL7/8gsTERK2yPKz6zJfafn91aZbF1NDQEOvWrUNqaiq6du2KN954AwsWLNDYd968eZg3bx68vb1x4MAB/PTTT7Czs9NqetOmTUNUVBTefPNNeHl5ITExET/99BM6deokxcdpUMZ58+Zh+PDhePnll+Hr64vMzEzs2LEDLVu2rNYvMjISfn5+yM3NxdatW1X/O8+aNQu+vr4IDQ1F//794ejoqPGMnrFjx+L27dsICAjA5MmTERkZiQkTJjTos7Zp0wZz587FzJkz4eDggClTpqjes7GxwfDhw2FpaVnnmUXm5ubYsWMHCgoK0KNHDzz//PN46qmn8OWXXzYoV2MwMjLCli1bUFxcjB49euDVV1/Fe++9BwAwNTXValz1/f7rYm5ujn379qFdu3YYNmwYPDw8MH78eNy5cwfW1tYAgOXLl+P69evw9fXFyy+/jGnTpsHe3l5tPJs2bUKPHj0watQodOnSBe+8846k29Trq675Utvvry6P7G1LLl26hPbt2+Po0aOyvfVwU8jY2J566il4enri888/13eURpGSkoI+ffogMzNTbQcNyU+z3QFFzcv169eRnJyM5ORkfPXVV/qOozObN2+GpaUlOnXqhMzMTERGRqJ3794spE0Aiyk1Cd27d8f169cxf/582e5AkkJRURFmzJiBrKws2NnZISQkBIsWLdJ3LKqHR3Y1n4hISs1yBxQRUWNjMSUikgCLKRGRBFhMiYgkwGJKRCQBFlMiIgmwmBIRSYDFlIhIAv8Pom7W3r+D3sgAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 325x325 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "subplotpars_scan = matplotlib.figure.SubplotParams(left=0.125, right=.95, bottom=0.25, top=.975)\n",
    "\n",
    "for key in scan_dicts:\n",
    "    filename = f'{data_cache}/{datetag}_scan_{key}.json'\n",
    "    print(f'Reading {filename=}...')\n",
    "    df_scan = pd.read_json(filename)\n",
    "    print(df_scan)\n",
    "    #for acc in ['TP', 'TN']\n",
    "    #ax = df_scan.plot.scatter(x=key, y=['TP'], alpha=.1)\n",
    "    #ax = df_scan.boxplot(column=key)\n",
    "    # https://pandas.pydata.org/pandas-docs/stable/user_guide/visualization.html?highlight=errorbar#visualization-errorbars\n",
    "    fig, ax = plt.subplots(figsize=(opt_scan.fig_width, opt_scan.fig_width), subplotpars=subplotpars_scan)\n",
    "    for method, color in zip(['HD-SNN', 'corr'], ['b', 'r']):\n",
    "        gp_scan = df_scan[df_scan['method']==method][[key, 'TP']].groupby([key])\n",
    "        means = gp_scan.mean()\n",
    "        errors = gp_scan.std()\n",
    "        # means = np.array(gp_scan.mean()).squeeze()\n",
    "        # errors = np.array(gp_scan.std()).squeeze()\n",
    "        means.plot.bar(yerr=errors, ax=ax, capsize=4, rot=-60, legend=False, color=color, alpha=.5)\n",
    "        # ax.errorbar(x=scan_dicts[key], y=means, yerr=errors, color=color, alpha=1, label=method)\n",
    "    # ax.set_yscale(\"logit\", one_half=\"1/2\", use_overline=True)\n",
    "    # ax.set_ylim(0.1, .99999)\n",
    "\n",
    "    ax.set_ylabel('Accuracy')\n",
    "    ax.set_xlabel(label_dicts[key])\n",
    "    #ax.set_xscale('log')\n",
    "\n",
    "    ax.set_ylim(0, 1)\n",
    "    #fig = ax.get_figure()\n",
    "    if not figpath is None: \n",
    "        printfig(fig, 'THC_' + key, figpath=figpath)\n",
    "\n",
    "    pos = ax.get_position()\n",
    "    print(pos)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f10e7955-232f-4622-af7c-1247a356ceab",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Learning Spiking Motifs: supervised learning"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "5e890acb-9ea6-4867-a4fb-219e3382f85b",
   "metadata": {
    "tags": []
   },
   "source": [
    "## testing inference with wrong SMs "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "f4247adc",
   "metadata": {},
   "outputs": [],
   "source": [
    "opt_learn = Params( datetag = datetag + '_learn',\n",
    "                    N_trials = 256,\n",
    "                    seed = 2373,\n",
    "                    fig_width = 4.5,\n",
    "                    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "e72e77f2-4bd8-494c-b142-31cdd9c630ab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Params(datetag='2023-09-27_learn', N_pre=128, N_SM_time=31, N_SMs=233, E_SM=4.0, p_SM=0.05, T_trial=1.0, tau_decay=0.4, tau_rise=0.05, N_time=1024, N_trials=256, p_B=0.001, p_A=0.001, seed=2373, verbose=False, fig_width=4.5, phi=1.61803, N_SM_show=5)"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env = ABCD(opt_learn)\n",
    "env.opt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "791747fe-e1d4-4675-be84-35278201df0f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy=99.99%, TP=95.03%, TN=99.99%\n"
     ]
    }
   ],
   "source": [
    "accuracy, TP, TN = env.test_model(SM=env.set_SM(seed=42), SM_true=env.set_SM(seed=42))\n",
    "print(f'accuracy={accuracy.item()*100:.2f}%, TP={TP.item()*100:.2f}%, TN={TN.item()*100:.2f}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "28dd1ac9-fdba-4822-a1da-002e2a094143",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy=99.81%, TP=0.23%, TN=99.90%\n"
     ]
    }
   ],
   "source": [
    "accuracy, TP, TN = env.test_model(SM=env.set_SM(seed=51), SM_true=env.set_SM(seed=42))\n",
    "print(f'accuracy={accuracy.item()*100:.2f}%, TP={TP.item()*100:.2f}%, TN={TN.item()*100:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c7695cbd-68c1-44c7-818b-04269e117d2b",
   "metadata": {
    "tags": []
   },
   "source": [
    "## learning "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "65d8c083",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "999.0"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(1-env.opt.p_B) / env.opt.p_B"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "8a37a8bc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'data_cache/2023-09-27_THC_DEBUG=1'"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DEBUG = 10\n",
    "DEBUG = 1\n",
    "def cachepath(data_cache, datetag, DEBUG):\n",
    "    return os.path.join(data_cache, datetag +  f'_THC_DEBUG={DEBUG}')\n",
    "cachepath(data_cache, datetag, DEBUG)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "cdca9283",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Params(datetag='2023-09-27_learn', N_pre=128, N_SM_time=31, N_SMs=233, E_SM=4.0, p_SM=0.05, T_trial=1.0, tau_decay=0.4, tau_rise=0.05, N_time=1024, N_trials=256, p_B=0.001, p_A=0.001, seed=2373, verbose=False, fig_width=4.5, phi=1.61803, N_SM_show=5)"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "env.opt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "5c6a41fa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor(0.201)"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target = torch.ones([10, 64, 100], dtype=torch.float32)  # 64 classes, batch size = 10\n",
    "output = torch.full([10, 64, 100], 1.5)  # A prediction (logit)\n",
    "# pos_weight = torch.ones([64, 100])  # All weights are equal to 1\n",
    "criterion = torch.nn.BCEWithLogitsLoss() #pos_weight=pos_weight)\n",
    "loss = criterion(output, target)  # -log(sigmoid(1.5))\n",
    "loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "769446ad-8535-41d5-9f82-f33b9a8531b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "SM_true = env.set_SM(seed=env.opt.seed)\n",
    "\n",
    "class Net(torch.nn.Module):\n",
    "    def __init__(self, N_pre, N_SMs, N_SM_time, bias=True):\n",
    "        super(Net, self).__init__()\n",
    "        self.N_pre, self.N_SMs, self.N_SM_time = N_pre, N_SMs, N_SM_time\n",
    "        # https://pytorch.org/docs/master/generated/torch.nn.ConvTranspose1d.html#torch.nn.ConvTranspose1d\n",
    "        self.conv = torch.nn.Conv1d(self.N_pre, self.N_SMs, self.N_SM_time, \n",
    "                                              padding=self.N_SM_time//2, bias=bias)\n",
    "        self.sigma = torch.nn.Sigmoid()\n",
    "\n",
    "    def forward(self, a):\n",
    "        # N_trials, N_pre, N_time = a.shape\n",
    "        b_hat = self.conv(a*1.)\n",
    "        # N_trials, N_SMs, N_time = b_hat.shape\n",
    "        return self.sigma(b_hat)\n",
    "\n",
    "lr = 0.0001 * DEBUG\n",
    "lmbda = .01\n",
    "N_learnsteps = 2**10 // DEBUG\n",
    "# pos_weight = (1-env.opt.p_A) / env.opt.p_A\n",
    "\n",
    "def learn_model(path, env, SM, SM_true, N_learnsteps=N_learnsteps, #pos_weight=pos_weight,\n",
    "                #N_time=1000, # N_time_blur=200, \n",
    "                lmbda=lmbda, p_B=opt.p_B, do_stochastic=False, max_quant=10000000,\n",
    "                lr=lr, seed=None, seed_offset=5, do_supervised=False):\n",
    "    if seed is None: seed = opt.seed + seed_offset\n",
    "    torch.manual_seed(seed)\n",
    "    \n",
    "    N_pre, N_SMs, N_SM_time = SM.shape\n",
    "\n",
    "    # https://pytorch.org/docs/master/generated/torch.nn.BCEWithLogitsLoss.html   N_pre, N_SMs, N_SM_time\n",
    "    criterion = torch.nn.BCEWithLogitsLoss() #pos_weight=torch.tensor([pos_weight])) #  * torch.ones((N_SM_time, N_SMs))\n",
    "    # https://pytorch.org/docs/master/generated/torch.nn.BCELoss.html\n",
    "    # criterion = torch.nn.BCELoss()\n",
    "    criterion_reg = torch.nn.BCELoss()\n",
    "\n",
    "    model = Net(N_pre, N_SMs, N_SM_time)\n",
    "    model = model.to(device)\n",
    "\n",
    "    SM_true = SM_true.to(device)\n",
    "\n",
    "    if os.path.isfile(path + '.pth'):\n",
    "        model.load_state_dict(torch.load(path + '.pth', map_location=torch.device(device)))\n",
    "        # TODO: record during training : df_train = pd.read_csv(path + '.csv')\n",
    "\n",
    "    elif os.path.isfile(path + '.lock'):\n",
    "        # we want to have a file but it's locked\n",
    "        print(f'Path {path} is locked')\n",
    "        return None, None \n",
    "\n",
    "    else:\n",
    "\n",
    "        #optimizer = torch.optim.Adadelta(model.parameters(), lr=lr)\n",
    "        optimizer = torch.optim.Adam(model.parameters(), lr=lr)\n",
    "        print(env.opt.N_SMs, env.opt.N_pre, env.opt.N_SM_time)\n",
    "        for i_step in range(N_learnsteps):\n",
    "            # draw causes (SMs) - hidden to the observer\n",
    "            b_true = env.get_b(seed=seed+i_step).to(device)\n",
    "            # generate the corresponding raster plot = what is observed\n",
    "            a = env.draw_a(b_true, SM_true, seed=seed+i_step+1).to(device)\n",
    "            # print('A', a.min().item(), a.max().item())\n",
    "\n",
    "            # learn the SMs\n",
    "            optimizer.zero_grad()\n",
    "            if do_supervised: # supervised learning\n",
    "                b_hat_bin = 1. * b_true\n",
    "            else:\n",
    "                b_hat = model(a)\n",
    "                if do_stochastic:\n",
    "                    b_hat_bin = torch.bernoulli(model(a))\n",
    "                else:\n",
    "                    if len(b_hat.ravel()) > max_quant:\n",
    "                        ind_quant = torch.randperm(len(b_hat.ravel()))[:max_quant]\n",
    "                        b_threshold = torch.quantile(b_hat.ravel()[ind_quant], 1-env.opt.p_B)\n",
    "                    else:\n",
    "                        b_threshold = torch.quantile(b_hat, 1-p_B)\n",
    "\n",
    "                    b_hat_bin = (b_hat > b_threshold) * 1.\n",
    "            \n",
    "            # TODO: make a threshold per SM\n",
    "            b_hat_bin = b_hat_bin.to(device)\n",
    "            a_hat = env.model_a_logit(b_hat_bin, model.conv.weight).to(device)\n",
    "            # print('A', a_hat.min().item(), a_hat.max().item(), a.min().item(), a.max().item())\n",
    "            # print('B', b_hat.min().item(), b_hat.max().item(), b_hat_bin.min().item(), b_hat_bin.max().item())\n",
    "            loss = criterion(a_hat, a)\n",
    "            loss_reg = criterion_reg(b_hat_bin.mean(axis=(0, 2)), p_B*torch.ones(N_SMs).to(device))\n",
    "            loss += lmbda * loss_reg\n",
    "            \n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "\n",
    "            # b_hat = b_hat.cpu()\n",
    "            # b_true = b_true.cpu()\n",
    "            # b_hat_bin = b_hat_bin.to(device)\n",
    "            if i_step % (N_learnsteps//64) == 0:\n",
    "                print(f'Train Epoch [{i_step:06d}/{N_learnsteps:06d}]\\tLoss: {loss.item():.3e}\\t(reg: {loss_reg.item():.3e})\\tAccuracy: {torch.mean((b_hat_bin == b_true)*1.):.3e}\\tTP: {torch.mean(b_hat_bin[b_true==1]*1.):.3e}')\n",
    "\n",
    "        torch.save(model.state_dict(), path + '.pth')\n",
    "\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "2755507b",
   "metadata": {},
   "outputs": [],
   "source": [
    "subplotpars_learn = matplotlib.figure.SubplotParams(left=0.125, right=.95, bottom=0.25, top=.975, wspace=0.05, hspace=0.05,)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "db83c85f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 450x278.116 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = env.plot_SM(SM_true, colors=colors, N_SM_show=4, subplotpars=subplotpars_learn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "f7eae9ab-5dc3-4012-80c0-6b11956e3f54",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "128 233 31\n"
     ]
    }
   ],
   "source": [
    "SM_start = env.set_SM(seed=51)\n",
    "SM_start = env.opt.E_SM * torch.randn(env.opt.N_pre, env.opt.N_SMs, env.opt.N_SM_time)\n",
    "N_pre, N_SMs, N_SM_time = SM_start.shape\n",
    "\n",
    "model = Net(N_pre, N_SMs, N_SM_time)\n",
    "print(N_pre, N_SMs, N_SM_time)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "5fbdb0e9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcUAAAENCAYAAAB+en5dAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOx9f1yV9dn/BRzwCIcjIikSEhEjYmRERIyRMR5mzpwxYmZI6pg555z5OEdmZOjjzJg558wxc2TmmDNizDEfxswRkSOGRIwRMSIiQkJSQCLEA3z/uO/7vN+3nGP+aF90z7leL1/enHOf+8f1+XVd7+t9XR+nkZGREXGIQxziEIc4xCHiPNYP4BCHOMQhDnHI1SKORdEhDnGIQxziEFUci6JDHOIQhzjEIao4FkWHOMQhDnGIQ1RxLIoOcYhDHOIQh6jiWBQd4hCHOMQhDlHFsSg6xCEOcYhDHKKKY1F0iEMc4hCHOEQVx6LoEIc4xCEOcYgqjkXRIQ5xiEMc4hBVxnRRLCsrk29+85vi5+cnTk5OUlhYaP3u3Llz8thjj8mtt94qHh4e4ufnJwsXLpT29nbdNRobG+X+++8XHx8fMZvNEhcXJ3/961//P7+JQxziEIc45D9BxnRR/PTTT+W2226T5557btR3/f39Ul1dLU8++aRUV1dLQUGBvPvuuzJ37lzdeXPmzBGLxSJHjx6V48ePy2233SZz5syRjo6O/1+v4RCHOMQhDvkPkTFdFD08POTtt9+WH/zgByIi8uabb1q/c3d3l8jISPmf//kfiYyMlOTkZJk0aZIcP35cWltbRUTkD3/4g/zrX/+Sw4cPy2233SYhISGya9cu6e/vl1deeeWin2NkZER6e3vFURv9/6849D424tD72IhD79eGGMby5hUVFdLQ0CADAwMiokChmvT09Mjvfvc7GR4eluHhYRkcHJTy8nIREfHy8hIRkVmzZslNN90k06ZNk08++UTeffddGRoakqGhIXnooYfs3vfs2bNy9uxZ69+9vb0ybdo06Xn+eTG7u4sEB+PkNWtwrH1eU4PP9u61HlYOTLcem0w4JWxJLP7YsgXHN9yg/Ps/IA8/LLJnjx29v/++mM1mKavztn6ndgmZGYg+oVPqkiXWw+pNh63HDQ04JVXyrMeRW1Nx/u8/+D+jdykvl7N33mlT7ydP9ojZbBa3I9CfWCzK/xUV+GzTJuvhMNnRziuW45yYGByr41NE0JAiklo4T/LQJP/RcvKkiNlsu7+/9FKPuLubpaVF+Xx11zrrOY2LN1uPGeyaUbrRelwWvx6fxw1bj0uOoG1mVuE6kpgoEh19Ja9z7Uhzs0hQ0BVdYkw9xVtvvVUefPBB+fWvfz3qO1dXV/nSl74kW7dulZqaGnnllVfk3LlzIiLS3d0tIiLjxo2TjIwMef3116Wurk6GhoZkeHhYvv71r4u3t/eoa2ryP//zPzJhwgTrv2nTpv1b3u+K5eRJke9/XyQgQGTcOBFfX5F77xV54w2cExgo4uQkcuDA6N9/+cvKd7RwX1CGh0XMZhHNOAkJESkr059TUCAyc6bIpEnKtdlA+By5VvR+8tOT8v2i70vAzwJk3KZx4rvVV+7df6+80Qq9B24PFKcNTnKgbrTev7zry+K0wUn21uy9qPsNjwyL+WmzNH6i6D3kFyFS9gH0fm7onDz2l8fk1l/eKh6bPcTvWT9Z+PuF0n6m3d4ldXLN6P0q7O5ZWSKhoSIeHiITJyrrCwFaF5RrQe+DJwel8fuN8reAv8lr416TN3zfkLfvfVt63uixnvO3wL9JqVOpfHzg41G/r/xypZQ6lcqJvScu6n4jwyPyuvl16W/sFxGRN0PelO6ybrvnv7vsXSl1KpUPt394aS92BTKmnqIGn+bm5oqI6Eg0DJ++9957Mjw8LC4uLiIicubMGRFRyDjf+973rL8ZGhoSEcUDPXHihEydOvXSHmjKFKX3E+Enew5GSUqK8n/QKoprapa1iETvheV8NGWX9Tgn6pj1eIcvPJ+j790gCRdyWB54QGRwUOTFFxXr5+OPRV59VeSTT/TnTZsm8sILIvPn47OKCsXU9PC4wA3Ok7o6EaNRmR0+/ljkgw9E7rxTf86nn4rExYnMmyfyyCMXfemHHx494WgyaPKWQZNZIiLwmbn4oHIQmGT9LHe/m/U4MQfezUAbfpca04w/TInWwxxy/ve/foOkXUDvDxx8QAaHBuXFpBclaGKQfPzpx/Jq86vyyWd6vU8zT5MXal6Q+eHQe0VbhXT0dYiH68Xrva6zTowGo4RMCpGP+z6WD3o+kDv9oPf+c/1S3VEtT854Um6bcpucHjgtjxY/KnN/O1eqllZd+OLqmLElbkUF4ubuLo0R86yfaUDGnDno43378ZucHBwHBqKPZ0bh87Dy3fiDXPc8r1IRwW/Ol6uxu4eEiOzcqTzPZ5+J/Oxnik3Y1CRy3XX2r30BtYuHh/Jv9QHFezu6pdL6nVcfzlNtf0U0t1JEfHzo86ws6+HMtWutx4Px8D7dairFnvzzgX/K8OCwhL4YKuODxsvgx4Ny+tXTcu6Tc7rzxk0bJx0vdMiU+VOsn/VU9Mhgx6A4e1y8b/Vp3afibHQW9xB3Gfx4UAY+GBDPOz1tnnvy9yelt6JX3PzcbH5vU44du2JP8aqCT0+dOmX9ToNPh4aG5OzZs+Lk5CSDg4MiIlaLa/duZfDdd9998v7778s777wjIyMj4u7uLi+++KKspU7CkpGRIcuWLbP+febMGQkLC5OaCfeIyWSWyBj0zAzTUfxwkzo7UEc8FRhpPT4SjwFfcwQ/Y6s2Li7Eejzvjw+LJLxk8xmlu1vk9ddFSktF7rlH+eyGG2zDIAsWKKP1ww+VGUNEJDdX+XzfPtvXtyXHjonEqlBvebnI7beLjB+vP+fhh5X/aZBejMy89YTExNjW+/e/L+Lqqoc+9+5VJurpMmj9LN2/BCeYMAsX1QAVqBEMiOU1S63HA2mYqNO8ikRkjs3n7B7oltdbX5fSRaVyT6Ci9xu8bpDo60frfcGtC+RnFT+TD3s+lGkTFL3nvpUrC25dIPvevni9H/vwmMROU/Re3lout/veLuNdofcJxgnyl4f/ovvNzm/slOg90dLa0yoBEwLsX7y7225/l3feERk3TowxWBQ1w2QOqcdtWbr1OHR7rvWYkFEJ64bhxx2+8wDGz+Rx8D5sPOZV2d1TU/V/b9sm8utfi9TWivzXf9m/tnd/m129f+UriofaX6osVglGQKDSBgsvMpRWP1Oa9bCrCx9PLwSs2rYTny9eTM+8MczmM57rPic9r/dIRGmEeN3jJSIixhuMYo42jzp3yoIp8uHPPpSBDwfEOM0oIiIduR0yZcEU6dh38aTGnmM9Yo5Vrt9T3iOm203iMn60BXH2o7Pyrx/+S277821Se1/tRV9f3C5hAbUjY7ooavDpHXfcIcnJybrvXF1d5aabbpIzZ86I0WiUiRMnSlWVYhV3d3eL2WyWFnVifuONN+T222+X+vp6ERE5efKk9PX1iT3Ztm2bbNiw4d/zUl+UmEzKv8JCJV4zbpz9c6dMUXCmF18UycwU6e8X+d3vRF577eJmCS0GNDAgMjKi/H32rMjQkHIcFydSVHTFr3Qt6N3kZhKTm0kKGwolxj9Gxhns632KaYrcG3yvvPj2i5I5I1P6z/XL7/75O3lt8WsXtSh6bfESEZEBy4CMyIh4bfGSs0NnZWh4SLy2eElcQJwUpdrWe8/ZHnESJ/Eyen3ufa4JvV8D3X1wUGT3bpEJE0Ruu+3z73O1693F5CIuJhfpKuwSc4xZnMfZ9/hcp7iK973e0vFihwRmBspQ/5B0/q5TIl6LuKhF8XWv10VEZHhgWGRE+Xvk7IiMDI3I616vy4S4CTK9SOFkjAyPyDsPvyMBPw4Qjy9fguv/BcmYLopOTk5SVlYmv/rVr0RE5PTp01JTUyPe3t7i4+Mjra2t0tLSIkNDQzIyMiJGo1EGBgbk1KlTEhAQIMEq8aW7u1uXmzg8PCxOTk5277t69WpZQkQNzYKLeOsFMY8fr2AjmrDJrHmIRPjwzlppPZ4XGorjLFz/6affsR77+2M0PRn4kvyPvYc0GJTgyCOPKJhVZKRiQs+fLzJ9+ujz09NFfvQjkSeeEMnPF7npJtHhkReSmhpldrjjDpG8PCWIMnOm8r6xsQrGdKXS2GhX77+2LBKzk6vImiR6JhWW3gqXOzceM156GzwWHx94Muy95MXDOywi2G/Gxz/TtyuJwdkge+/fK4/88RHJOZ4jkVMj5Z4b7pH54fNl+pTRek+PSJcflfxInrj7Ccmvz5ebJt4kEb4R9rSgk5plNTIyMiJ37L5D8h7Ik1CfUJn50kzJis+S2GmxYjTY1vuAZUAeO/KYPHTrQ2IeN9qqZ3mx8xuyevVXbep9Qf0T4upqlgJjp/W7lcWLRURktxHw9NJVq6zHoYG4dmYmvUswyGRFBniHf5kC7DM/f5I88IDt57yau3tRkfIc/f0iU6eK/OUv50GYNuRoo7/d/j4woDg03iYVBWlqsZ7TagSStIU4fmlpCdbj8lJ8XptVYD1uj4FjwZy0xStMNuOszgZnCd0bKu8+8q6057SLKdIkXvd4yeT5k8U03TTq/KnpU6XpR01ywxM3yMn8kzL+pvHiGWEb+jxfomqiREZEjt9xXG7Ju0XcQ92ldmatBGYFijnWLM5GLMitz7SKk8FJrl95/UVdWyff+tal/+Y8GdNF8fe//728/vrr1r9Pnjwpt99+uyxatEhWrlwpTbw4iVhh1o8++kgiIiLEbLY9ITg7O8vw8LDN70REsrOz5Sc/+cnoL+6+W8RkkpJAwG4zt6OjbYtTOuBqH0zOu8N3WI9pTZQusjRH/vSR9bg3HIviyZN2H1GRBx4Que8+BVeqqBD53/8Vyc4W2bNHj4+IKOd973tK4C43V5k1LlYCA0UqK0Xc3UVmzVIgnPZ25f4XMtkvQf7QfY/8PTvTtt4ffFC5Ny9UakBl9oGF1o8OH5iJ73cCK5oXCIj1WBXgk1gLgpipEWBT/nbaq2KfmyzyQNgDcl/IffL6B69LRVuF/G/T/0r2G9myZ+4eWRyxWHfufSH3yfeKvidlH5RJ7lu5kn77xes90CtQKj+qFHdXd5kVPEvaetuk/Uy7PHDLA3Y91HND52Tey/NkZGREfnnfLz/3Hosm/69kZr9hU+93360uACqrW0QUN0n0k6okJVkPzcXF1uO1azGB+1Vgcp61E2OmpWUSHV/4Wa/W7v61rykLaVeXyPPPK+H0N98UmTzZ/j0S4oclc73tecbba1jM5mFp71D6qsELesyn+O2uTCJSUWwhtg+GYmM4WKYhtHC274dhMnu2/ee87oHrxPs+b+l5vUd6K3rl1P+ektbsVrl5z80ydbGek+F9n7cMfW9Iesp65ETuCfFN97V/4fNkfOB46a3sFWd3Z5k0a5IMtA3I2fazct0D1+k81DPHz0jbz9skqjrqgo6NXWlqErnllkv/HcmYLor33nuvDA0NSVhYmPzoRz+S6667TkpKSsTb21vc3d3Fx8dHenp6xNfXVyZNmiQ1KtPx5ptvFhERNxU/DgsLk/r6etmwYYM89dRTMjw8LFOmTLF322tLjEaRr39d+ffkk0oqwlNPjZ4lDAYl3vfUU8qI/f3vL+763/iGMgtZLMo/k0nBkc6eVRimIiIXgKL/U8VoMMrXb/q6fP2mr8uT9zwpSw4tkadKnxq1KBqcDfLw9IflqdKn5M2P3pTfP3hxev/Gb74hr3/wuliGLWIZtohps0mGRobkrOWsTMpW9N63Tq/3c0PnZF7+PPmg5wM5uvDo53qJ16Jcjd3dw0PJxgoOVqDdL31JiSs+/vgVv+5VIS5GF/H+urd4f91bAp8MlIYlDdLyVMuoRdHZ4Cy+D/vK+0+9L2fePCPhvw+/qOvXfqNWul/vlhHLiIxYRqTMVCYyJDJ8dljKJykG2Yy+GSIi0v16t5zrPCd/C/gbLjAk8t6P3pO27W3ylZavfDEvfQEZ00WxqalJl46heYopKSmyceNG6VIjyh9++KF8+CEouaWlpRIcHCwfffSReHp6SkNDgzg5OckWygGM4byp88Qu8eCjj0Q8PMTfH5Yb4w6rVOs5j7wXdmaXBhMppwGeyTYjmGCriwG3Wu6Bl3nREhamY8fqJD1dZOtWxfOaOPHirrdnj0KrW7RIJDlZ5P77ldzM0FBdLuCVyv2eR+VrdvReMDBb3J3NMm8JzP2S+Qo8ejiQcuGSgCfllqONDJRStzAQ3uHBjhnWY68IHD8UdUpE7Kfs2JKw68KksKHQ5nfpt6fL1r9tlQe//KBMHH9xet/zzT3ymeUzWVS4SJJDk+X+0PtlTckaCfUJlSWRo/WuLYj/+uRf8tdFf5VJ7pNsXNWGvPmm3f7e16eSpwkL3Oev9NWFNRm4Bv223gK9+xLMuLIU3uEyQkwYPZk5aJ/gZE+uxu4+PKwsoheUM2fs6r2p2VlMJmfxVR0tTgld7X/Qetw8AAJUUBVcyJJ4eIele/Fb3yTMP8sI3p037W8icvGLiUeYh3QVdtn8zjfdVz7c+qFc9+B14jrR9aKud/Oem2XosyFpWNQg1yVfJ5PunyTvrXlP3EPdZeoS/cLr+7CvTEzUN2btvbUy5eEpMvU7n59NcPAft8i8K3MUx3ZRtFfZwcPDw5qTaEu0VIujR49a0zNERD777DPr8bZt2+SArWQmuQB8ejXJJ5+IfPvbysifPl3E01OkqkrBk+6/3/ZvbrlFwXjc3S/+Ptdfr8yMtbUi+/eL3HijcvzYY/oiBpqcOiXS2qrgTSIi776r/O/rK9ZRbkeuBb1/0v+JfPvlb0v67ekyfcp08XTzlKr2Ksl+I1vuv9m23m+57hbp+nGXuLtevN6vN18vlmGL1H5cK/u/tV9unHij1H5cK4999TEJ9tbr/dzQOUl5OUWqT1RL0UNFMjQyJB19CrnBe7y3uLlcmHF3Tej9Kuzun34q8pOfiMydq8QSu7pEnntOsZ2//e3Pv9fVrvdzn5yTf377n+Kb7ium6SZx8XSRM1VnpDW7VXzutx009bjFQ77a9VVxdr/4NIxx14+TYcuwfFr7qdyy/xYZf+N4+bT2Uwl4LEDcg/WN5zrJVVwn6RdbJ1cncfN1E/ebL6Ghr0DGdFG87bbb5O6775Z33nlHurq6dPCpj4+P3HTTTdLc3CwjIyPi6uoqFotFRkZGrPDpokWL5IYbbpCKigo5fvy4NU/R3d1dfvrTn176A338sYi7u4QZKKHO6G897OhT4CoOfUVRflZuOTD99GUR1uP4FpyzoxzeofG8/CudmEwid92lcM/fe0/k3DmFf/7IIyLr1tn/3aSL9CBYqqoU2t2NNyoBlo8/1r8Yy6FDIt/5Dv7WksWeekqXqnK+vD0pQUSO2vwu+SsnxOzZJ2JMsn6m3b7aB2kukVnInUvfGmg97vWF99LcBY9wngX3yyhG28yc8qGIneIOJjeT3HX9XfKzip/Je6fek3PD52SaeZo8EvmIrLvbvt4v2nMjqWqvEi+jl9w48UZp622Tjz/9WKL8Ruv9ozMfyaF3D4mISMSvInTf/XXRXyU+MN7+Te66S5/9TrJmgar3BuTaLvRV0l56k7Ktn5n3bLMehzUdwgX2Iha5w4BrSB8lEHph/Ig6bm3J1djdXVyUUN6LLyoL4qRJSh7j668rhQIuJB90T7D73VNPKSlIGiTMi/GpKHiHA0zqVGO9IiJ99PnmxVTxaT+8yX4LUjXcPG2TYVxMLmK+yyxtP2uTz977TEbOjci4aePE7xE/CVhnP83n/EXrYuRM1RkxeBlk/I3jZaBtQAY/HhTPqIsj6VyKBAZe+TWcRsawEF96erq88MILoz5PSUmRxx9/XO644w6bvyssLJT7779fqqurJT4+XuctanLs2DH5yldsQwa9vb3S29tr/dsKazT1iKenWSab+q3f5R6AdZJuVGtUESY0Nwt5iofmo4bV9C1IcKqNAwQ4rwuTvLv7xVffuOblb3+T3i9/2abee+rqxKy5BppokwDPGIQzRS9G7lV+Pk5hVqB7V6vtL1xdlX//FyQrS3pXr7ap94KCHvHwMOuqskUHKkzUfhNYJJxE7rcTK1T0EcB4lZm0WDJLhymhb76pBPX+L8jOndK7cKHt/r5zp5jHj5fOOUq4YHJXPX63dav1cDAHDGu37TBS+lcA2nYvpRJ98fE4Zhr2n/8scoGyl/9JcvLkhYsqXIyMaZm3m+1Yjh4eHhcs06axUEdGRuTMmTPyla98RSZNmmSteCMi0kZJsOdLdna2TJs2zfovLMx2cqtDvlhx6H1sxKH3sRGH3q9NuWrh06lTp0pUVJRUV1fLyMiIjIyMiI+Pj3R1dcl/qaUknn32WRERqampkcHBQRkaGhJ3d3fp7++Xt99+W759McA/SW2twjSbGQx8Ij2cAs5tCrOg0x/eIZMAmlvIO5wPK7p3BbzDgxZU7VGsQirc+x8sv6z5ioj8yfaXjY0iHh6SvB9kjYJ4RU/pKbC0c/1xTDWqdUaxe7Cf9XjbGlDaZ83COa6uCoPw/4T893+L2Akl/NdQiZgt7iKbkMQ5XKh4fIyGcgUV01ryDgdWW49rAwGxcobHcn+C967C2p//NsnP10GeOlmwQMRslskNiod4uAWLZVc8vMNAIuDMIMTEfRmIfu1bkB7mtwJEteVGXGfX0xfIyfgPk8uB08+XqxY+3bx5s4SEhNj4lUhJSYl8/etfl69+9aty7Ngxm+eEhobKO++8Y/M7e/Bpz6uvitlk0kE+XG9TC59x/IHDaHV1ON64FhAsw4KH+xDz8vH5v1O8/oLwaUmJmD089FRFTdlUW7Z1L2KEAS0U9yVUoDIYhkm0oRrnULChudv7SssjXjvyr39J75QptvX+xhtiNpmk3oDseA0qpTRQydtCMDTpusyChH1Gp8N8UAzgcBVgWINBSZL/PyFtbdJrNtvWe1OTmD09ZfMeRTfr5qNe74zF6JjMXchIpL68Zw+OKX/x8BrKTRxA3qhER4v4U2z3P1mGh0WcrwwAHVNP8ULwqesFYj7ad14cDDlPtDqptuRqL7/0nyoOvY+NOPQ+NuLQ+7UpY+opFhcXy+bNm+3Cp1/96leltrZWzp07J8PDw7JgwQLZunWreHt7i5ubm+Tm5sojjzwiLi4uo1I47rrrLqngBCASe55iTY1CtDGQqaAjbsxRGYy8JyKRCo51EwxCkNPcCFjaeeVgdZ06JbJixQVV9B8ja9aIrF9vW+/vv6/s68d6Ny9Oxg81YSIOm9G0fRVDsNxMXE0lNva8ii3/wXLqlIjBYFvvhw8rRBtG+TSU7ggVtOe6+ge30C4knKRLjZfXAaYv7wlYX693cv6TpblZxMfHDiKlEcu08nkEN+XVwWtPjQL0XFAH1Iz78uo5BE9THOFwG64z+84vgH1yrcgHV75X6lVb5i0rK0v+/ve/687/zW9+I7/5zW/kr3/9q8THx8tXvvIV6ybE50vQBfAxe/lDN/6rRMzu7nLIAgyex/1qtUpwYxsYqRxTfOyxT63HI//8wHq8uxiL5dIuxGQ2DV+Aa/4fJltnlkhmdplNvX/6qYJ43HAD+sLIG+piSMHDxu1g2oVUIcnZ6UEwGkd+RdsWBS62Hvr7AwbfsEHkmWcu6zWuOUlPFwkPt93f//AHpQYnV3BprlNg/9370ceZQDp7BcbVli045jgiV59x30MFKrKSReT/BoxXXy9SUWFb7zWfTBPTWbNEahYCWSCpNVTnzSveehgaSqXgiG0t/jVCJ1kPZ8+iObH0HyIJMFT+o6W9/YoXxTFln957773y3e9+10qYue666+Stt96SjRs3yvXXXy+33367mM1m+YZK405KSpITJ05IrLrfyy233CJOTk7i5OQkmzdvlq9+9avWennzebM1hzjEIQ5xiEMuQsYUPs3OzpbHHnts1OcpKSny05/+VG688Uabv3vllVckOTlZKisr5a677hKz2Sx9fX1Wj9FsNstDDz0kObwjKoldos2HH4rZbJbcfNSUpFrI4n1gl/aA+JDMtpJg5CPyhvRMzGGr/I47RPz8lN3C/6NlZETkk0+k183Npt7r6hTYOqALZIJag8Lw5Ry5GU1g1AmV8WP2HkOwM4X2X+Qvpk9XaGr/4YofHlb2HJw40U5/X7lSzOPG6eubaXg+58uFAopza7MDnxK0rZWKE9Hvkbn5yc+U4qb/4XofGVEcFk9P23rPyOiRcePMsnGOsp/iMQvYdrEG2hCYyTEESTUmYp4JCadqRrTf1eYqMJrWfesdpXDCFRJQrnYZHla6pB1+5kXLVVvmLTAwUB555BF5/vnnR31fW1srycnJ0tio4Onc8bS/T58+bfe+V0v5paEhZfA4O4tM9aTVUitXx0QiDsbceiuusRK1VLnLO/35z9bjyvcRT4juxkKxpkQZOAx3hZ+hQry0mpf8FcQnTrfiHcZ/8Qsca3nEM6NOKb11ePiq0bsMDyt1xZydUa+Ld5PlggFcafqJJ6yH57ygU9dMGHbvLAYuW0Jr8scfK/8zt2yR6RX8QSUKXxmPDWUfmI3P/1CCHXDvd8EEeNQd8dXnnlP+//nPlf41NHT19HcZGUHcSw2M1Q2hUGW4gL4duRAQ2O9/j8onNwjCEo/twjnPLMfn8t57yv80Tk4K2uu68RhrlfUILnPbtCObR275NeLazcthLATlqdD+d79r/eyETJXh4atM7+fOibz7rjyYBV3vpijDhHzUn+ZqVQ8+hBnle9/DKQldCF2c/Boq8DzyiPL/yy/jXNd9dO0TJ6yHO72w79iKB7Fd0N+a0E5faXzRevy/kxeJiMg37qKUNrWCmYhI85nrrEXer1Su2jxFEZGvfvWr8o9//EOqq6tlcHBQYmNj5bnnnhNftcZmQIBCWtG2iuItoy7HAV66xixubmbZN5+qRAxE4FhdmKpjYKn5zMFxOa1bnDvHpA7aeUe/pV8dLPDlOYplvmsOPccPfmDzIn/KgncVTkXrg34Gaz2UcsukCxP+dJUEER5E6SNCu6fSRD2z7SXrcWcE8qGY+7I5BwUXStQ8w07L5xfentb8mpKSQW5hn6/iKc5Io3JTvPMrGQxMP68OBNGm0QRrmQkfMyqo+sr1yp5tBRMxsQ1gpy9JTcBu8dUfYsD6U/ea/Oij1uNbijDbjPsmtiALylQXXX+wWta/h76zMRNs6bu7ce2C/8VCyO0rAnM4wYKKKDEvKRaLu3F0nP18ObH8f6TP06xrw4g9JaPOy6f0jNJSxBHDw3EcEQFdL0zExNWeiPY/WIQ45bxgpW95U7ncfi+8IKeE3PAsDL+AQsQoW9PQx3f/GX08KkpZLDuIksBlkPcl4R2jKYH10BE839woWhVp7yUey2UzlIl9xutYJN4yYZGwJ/ffr16nTWHjWeiaJX3wGjcRx6zMC2O+2IJ+011um2Wv21tSNfJ+91v0ifoGLHgTHn4Y51LKzVNPYeyFVRBKQ+lNv6TdywrXKV5u/zm8w1+mYlzNfheVeFrJj6k9QQvhL5GDmWpBDqY1GkZ5bztqkN7G5d2utEbCVZun+PLLL0t8fLy89tpro77/0Y9+JFu3bpW33npLIiMjR30vIjJnzhz54x//aPM7u/Dp+++L2WyWwxUYyLPLMfAGs5SB59ZEZZmokdrjMCB48eMNO9j58ytGRyvwwkKT3K1+TgO2ugNJ6bzgcuGeeTFgudZ2o0Oz9cR5lYd81UmbPKMdRnTclVGUA0rFvncUYTJcmUQ5bESLq/VSOizDZ7NmXZgFSaUbrYzIhV60gDGcxMXHmT5JpIVmwXPSZgVSsh95dNps2ZqEiZfbiDecdauCPtoDkaPntww1Wauz8LyRyzA5dBYpE8ZkCybb7P1o04w1tIhxA/MxGQ3V3Xg3LmRQsEVBTxpp0fT1vTDrlxfFmTVKOTFdKbE0GBpLfWCA7I5B/81oQP/NXmYbYq30wsKp2TfslDMNgCd1XiATqH5uQTfII8nlKCRgtR744vSSzUk4l/vnbBPlvhLcvr4Ybc1s8cle6oLEGD9RdXu3b7c9zxw/ruRDa8/HVHVmLNF1d/RBvyuXkBFrjybM11RfsrUP8xrtGy2lpThm1GhbPPpyfyL6uHsV6YknFA2Gpz57iDYAn9uAUnV808gOGP/scOjiTtqqR/Gn4Tl4JucV9LtdKJZyOXLVlnkTEbvJ+ybVXDMYDOLi4iITJkyQqVOnyl/+8hfrORda67dt2+YovzQG4tD72IhD72MjDr1fm3LV5ikGBATIiy++KDk5OTbhU19fXzl06JAsWLBAPvvsM3F1dbXWRBURyc/PlwceeMDmfe16ig0NSv4Q51kQ7ladpFSeZ89LK6AsItLch+odQUkgJ7DFU7YY1nUmYHXd55rXmGyApXYqDlaRdx1Zatu345gtViIZzVwB46JkDlHkVQp3fxwsePcKWOIlFlji/KyV28mDpETO6Sm4j+YJcK7gli129K5W+Gi3QH+at+beVIsLsFtMONbuBsAoS6sAWbZnAcr08yGYifTUa1TuyemQ3L65mwhGY5eFsW+CAhqbYGcy1ObXp8a/aUcPcze87MN18OzZc1vvT7AV6Tp5L/pDQRJgprJABX6a0QfruzcuzqbeFy3qETc3s+wOJgtecxWYEcYvkoZ4J1ezr++G18tQNTssDHFFW5Q+1B8BL4xTDbi4ETc7t40O9o3AcWqHUnKuIBAeIf9uXiDILNUGePORFiK5kK7ZK+e85YA6RcfHvACvxvrCS+718bGp99de6xGTyWwl46VnUYiA3TbqlJ058NAn19FuM6wo+u2OLsTItfbYvInQCPZIuV8zeYoRGIIjOgdAROS2iSxcLyIih2OwQ8fsKEJlqBE+U8MWIiKP/RBL0I40agN6n87FCnLBMDirqiCH7jMZ88jlyDUNny5evFhefPHFUd+LiNxzzz1SylojyczMtBkA11iQDOOtW0sdSYUnTq3FJMIwaWoTOsM203rr8WojufOE4zkRSWXkeI31uNVHgYQZGWG0cHYXJsH+FGDw7jvxXLNLAX8djkO8ZbMADtbGHI+DMAstQjTrrGsCDMKD63AxFgFek7XxFNCEAZx59KhNvfecPi1ms1m39Y3fWuW92hso+MADmYuZ8lYjhG/l7QHMlJpIg4bxUa09GHqyc71+L0z87pkE1zEJiqyHbQa0gbaWTO62nWzNcF2tBR4Fw3vzgqnUF8VX61PQ18La1HgZrUaZ9fU29f6Tn/SI0WjWQWnOherkSx0uuxwLV0Yowdm0Eu0uxsTOr7WyhfTErG3t+anzLTQiNrePaoDqdqXZgmchgqyuSTVVeu/caPsEbi+miVO7p5ejv+emwMCI3YQFULM7pwuNGWrHzLw82/39zTfFbDJJboXSzulx1Cf4ediiYCiYmdRUoMR9E8Y2hwu2NakGVGIiPqTBeigC/YeNC54XEvqo3XkyYqtXhVKHw+EQONdQn6XxvdkHtXL5nmx0sCGj2WWskgQvXDt6GcJolbSuXo5c0/DpD3/4Q7nuuuvEyclJt0OGk5OT3XQMhzjEIQ5xiEPsyVUNn37yySfy+OOPy0svvSQDAwMyceJEeeGFF+Suu+4SX19f6enpER8fH7nlllvkvffek/5+eAYbNmyQ9evX27zv55V5Y/iHDUwN3lkfQww98gxq98BEYaag8wAFxtkSZFOITSDVw21cAaiTLSgmfzAJomwVIBZGVSbX4HnXlwMq3RioWuO8Dxu7exR1P7wF1jBbkCsN5AUT3KKVmWKLta7ODnxaU6PA1iyqsg+Hw9uaHUwWNTcSudS9a+AdsHqZuMqEDs3bWBcK3WVUQKfZKwBxlrXAG9IVwN4EqOrYCuypyQQrDeXmdpndB89IB8cyjkhWeXs42s6vggo+Ey45HKFYzM5r4KH1ZmXZ1vtLL4nZ3V0Hd6z3VSDnjVHwDPZ1A6rlyELBfvRrrvLEqFtnC/V96jj9wUr/YFSGhdvr0Co75Jpw6g80fvJaFG+S22hmHD0HG8wMEzOGTs9a1g3Ph8ee9uw7kvB8G2mj8VWrbPf3kJAecXExS/0WVcek/9a1GE8MD3Nf5imEiTH8WtuWkW5URTAT3EoSEpGCIuQ6csiG++SpbvhPXD2TnU/N0ea5j9sgZABzSEETdMp8moAcqvLFHrGG2BBCtGsJPMXlSRTm8AOiczkypikZb7/9tt0yb3v37pVly5ZJPk0Qp0+flqSkJHniiSdk06ZN8uc//1ksFovcdttt8tlnn0kTdWR7O2SI2M8fOntW2VYo1kgufxcGzXqj0huqfTBRR+ZjlvOia3F71tRgwvANBvzDA587/aw1yqQcQvTzzX1gRzIDjifzTOpcvGPHZIoZxCyjbQrClR7tHIjJnmOHG0n3pYQ4cVhpMBSsL34HHxVCazUBCszMTrap94K3bhR3d7PMi6CBrDLZZgvo/YfKgRzMnRWI5yxFTDHTTk1TXvfdKhCT7aNdSzTJbsKi2GyBfnmy182rq7AQbiEm6KFCwMwWizKpMLTIbGUvur87L5AUz/PbCcipOh6LXqQX4ljahBlJ17DX3+uC5orJZJbpPph9N/oqk/Ip2hh7CaFlg3vxrtk7YQzwRNg5H31VyuldqB/mpyiT71IDwaQ0wy7NR0GB4XgYdcmlFE/zicAxMZ81wyOyipLx9sBwOpQGY2RuMDHJCT7dJXj/5f5YKHrjYSBoiHvyCiyEBXvQXzOzt9nU+9/vXC5mNzcRo2KdHU3BQthXg/OSE7GgDhoRx+N+yKj/vkyMn5U7MVZ2bFUWwMkpePa8+Xin1AjowOnLGA8jryFcUdyGzxnS5zixdswLIaO+gxYshMlNZNQVk0NA/d3pifvwLN3qwKI+FMdjnZ2KK1wUxxQ+1fIUfVQtcpm3kZERKSsrk7vuukvGjx9v/d5gMIi/Ss3XFsw//OEP0tysTAyeqsdhMIzpeu8QhzjEIQ65BuWqJdo888wzctNNN9n83W233SY1NTUSFxcnb7zxhs1z7rjjDqli/IHELvs0LEzMLi6ybg7c/M2ZoyEi9sI4dS66CwH5g30IyM8rRtCeYb+SVTh/ZjDldmkEIbpR/RJ4CextMNGHS9KF+VLlB7pn+lZ4bprHx9fTJcFmwmOqXAvLLrqFYD+GMQkr1SAZhka8vD4HPmWzV4WyskvBDsyIpwg60dAOxkA387oIziXkgPUXJrCMNU9WYxKKiAwmou1Yv3PDqY10bjFM4+hZgKgYZtK6IuuXkQImETMSsLSOvC7GXun+f6fiAXfefruIiJw6ArTD3i4ZL7/cI+7uZh0kqCHRm9fi/GETvBQmKjIMPTmHSC3cEZktRhBDpygMwck+8KZDQmGjN4aj7zEaws+luzbDoKqS1x+Bd7NxMbUddfjUTRgPDAWmr0E79rdhLLl3UV4uu0yqNAcDibG7S0ZIiJhdXODdEIWztgFQ5vRggny5IzJMQXj25EAgUp07MUb75yiIBPc3hvYDjERCYzIbtWNzi7PNUxZayNPX4AJuC5ogq/tCRp0qch7RLxiw7o4c6EIbYhzO4uL0uoF1lNCEy5CrlmhzgkoCnS8nTyplgW6lMk7ni+UC9X6ys7Md+UNjIA69j4049D424tD7tSlXbZm3N998U0SUhfPEiRPS29srN954o7z//vvWvROffvppycnJERcXFxmiOngiF14w7clvV/5Nxo83SxZvsLETVlm5lxLHYYsyoAHxjsZgeBhdZMS2ZsGaYja4pRTHMwOxiNdGKZ6lF90nrAge0OFAxDvWrYI12WuBpXiKgureBuDtjPFrTh7HKAKqCOsnznu0Dyze3lDEwsxVZJWRCXnkiBJnSTZSqTo7cmrCjWIxm8WbPEUtLUW343gH3qM6jbzD/YivVc7H520UC7OQd++biMmpQ7VSAyiW4VaKNi0nYlJLC1Xy2YK4RXURgvyV+eRJ9HlZD6ObFDO92QdxOEYZuCoNO98F8YgrJ2+i6k1Ehb/zzjvxuer2eRchbceezIzpFbNZdPGYmEzlHYPC4R02hyMWtZKCziU18OINcURqo/BOlz/iSPPWAjFpWaaMickp8dbPKijWK93oe0/fBE+mj3KBd6SRAtkIVjt53XZ8pEvhIUJN3nYKgJHH35sC79DcQnFHGhOHkpR34PZaWm6b3Mcy/ObfZdhstsbvN1OR9en5e63H9fPhfYcx5MIxZ4IYcpikQlO7e6LCY0gqRX6xWxf67Nwl6MuZhA41UU4ggxTMKZAqcvnUMXSwBf0iieaWyE2km3D0o+JieIR9cTheOR/tvnqL0u6cPrRtBfS2OoXG3RXKVQuf3nPPPfLDH/7Q5u/c3Nzk7NmzIiLi4uJicz/F5cuXy3NadeTzxB58qpUbm2HCRFwtmIi0BYXTH+e2YbHqTcNiZa6jBHfuUQR91M5BJ55ehYVzY5syeaxfhWesb8MkZYOoqpxvoBqnDO3Q/atn4Z4a4uBtIfiERzjN1GUrAMcwQXWuEQvI7hYsIEt9lUB+ZwwmVKPRtt5Pn1bKjTlXAR5t9VUGVoCJYGBiy+StBcTNz7O8Gzo4GoN3TQgk+IxFXZlaOzAYOQ2LhREshnE4tz2yCXoaToHxoCFk/DtzPtpcM4RE9PA8I5HuNdSnmAbKu1yoZJHa+dBDYKBtvZ84oeidUWttop68Fzmvg6tALGPCEpOy7CC7ElIEI4UT0g52KUYTE20PBlJubTzl3AbSosTUS+IN7BuArhdGKH2jNxALsvkAkW7oJdbvB6THcDC3acASIqdRZ2vcr/TXkC5qF8IFe0Vsw6cvvCBmd3fZ0aE8M0OZ0UfQbsvb0H95IYrNQW4yNwK/y8YajDsrgYixb2qLxlCcyygkE9L4IU/1YaxwX9UMfja816+B0V5WBaN9Rg6MQ13+KmH5u9vgZCzVjBR+B+73HAO6wF66FyNj6ileCD7lvMPzRVvHz507J8PDwzJ+/HhxcXGR/v5+cXFxkXPnztldUEWU8ksbNmy4sod3yCWLQ+9jIw69j4049H5tylWbp3jo0CH54Q9/KM7OzmI2m6W/v188PT3lk08+kfHjx0t/f78899xzsmLFCklISJCWlhYdA3XdunWylk1gErtEm+3bxTx+vOwzolTYwgGyMDXL1E6lid1dgB6WdsPSbUyCBcykWLay5nqRVaZacYdK4R3OFdsVJXpDAVWYo2ApLk8EPXvXABF9qIBvqyipGLyP4er98IyJ5S4F2wFPLN2EFI7dW2wTejQSxEFfEEXsFQRvalLyQyenUHqE6gkdKgcMPDefLGQ25dlNYXOfCAwHm/BebJg6N6l6YneT3TlbeLOINA8AcgoqB1TZOQvPyPrTHHf27Fld1Cxy113vW48ffhh7ijJqxh4kQ2FugcpzDXYzrG65cGFqJqyoHkFBC/SV7EUQOSUW5iXC002dQwQYrnLCdbm4UpCGQvD3/FIUZ1idg37N4yfbMLpYv4iIW5ESAlhajPG4+wDGEjfCxiK853ohshBBqafmAwHiNtP4JIwsBBRjvuidP9+m3sPClDzF2mK13cg72rgT/Z3BAHaQ+Bm4bXY1IDVkeTAQnNnbFU+X+w87Wdz3NsfQPMPtaOdhyhKhM+t+p+z6EmJVIGiP5APw7OuzgK6EHQDE2rkC17YW0mdFECTOaVl20tMvWsZ0Ufze974nu3ljL1UWLVokd9xxh6xU9wp0cnKSkZERMRgMYrFYrIvij3/8Y9nKtZ5UcXZ2lscff1w2caCGxF6Zt566OoUFSb18XhqggoMrlIVrVx0aYHk8QTs8ufDEymwxDiryCOcZVJVO2o6HJ1PunzzPMILADDF74QiN9cU5y7wGcBjGbKHFjy/O70nH7T4KdOVnwWKauXv3BfV+sAKL7bx4FdIlfTk98WPr8dAQ9nd0ziHGKb8gs+BYaLHUdMxMUS47xcnlDGEF+GDRSZgDWIjng80rsFg19imLVYgvLR60EJ+8AXsD/vV3GJLcvtweHAdm9qh2PhdNyM213d+feUYp88ZMwISOvFE3O9aAiTp2LwxGna5ppmajkuPvDD9rhom5kGKfrDwuqsk/JHxvXRsWK57YtfExLwVhFScX7K86MjTResxlCtm2Cqqh2DqNzd4liF9bxwQ1UuMcfL9vn515ZudOMY8fb53g89bAKNXF99sAyy7MQX4zL2jMKM3egvc9VoH3ih1QFs7NFVg01x0AtLxrGUIRy9Oof1K8snE+VhqewmxVTPTuoDmR2m54FXTD12DIdmYbsVn54qrw1nCRe2hnDO6LtNXX5ciYsk/vvfde+e53vyvPPvusiOjzFDnP8PHHH5f8/HyZOFHpzE7qzt3B6gzu4+Mjzs7O1r+Hh4et5zjEIQ5xiEMccrEypp5idna2PPbYY6M+P59oo3mKrq6ucu7cOaun+Nvf/lZSacd07TwRkTVr1shPf/pTm/f9vDJvjMxxAedTPgqMw5bQ+gNgMm5cAcJKbQdYctO7CH6yhb8oL41j1W3pjILFY8ORFBG9dzgjtNP2F3Sfjfl4Xg2tmr6CYEtyK5tXgCQRlAMIeHAToGG3wtH5UCIi7jGqJUrEiN6+PtswnlYQnKE0Fb5qDIS3HNIBiDm3Cc+cHoP2iF6M96sspNJPrECGSrX35UZnT5hMWh3MkwbYbbAClr5bMcFP7Nqp7JPOPniVkxeTRUuuzsr9gMR3BIN9urwBUDR7FOzBDtepumAmpdFoW++qh95ugIfulxI7+gb0Hq3B8DZYZTqyFruuzKRhl1R1x3P3A4lJ7wAEWhYHaJQfxbkL98k7gjGW6osxtqNOeUbW43AFSFz2dt0IKibEgRlFZKDzLifa67TWQbdhMYBpKyrshGkSE8Xs6irtexRmtp8B76Tlb57/CCz1e4jYw3Ai6XefP/S3cK+ij8Yc6ChEqHoU53NT+GHjJvhM61cAKRr2AnLAzTvPoHjX7THw5phhPWyBJ6u7NsPW1Hei8zHnVJaqyAxDJwy3cydptkOqu0gZU6KNvfXYw8PD+p2Tk5NMmDBBF1PUxM1NGVA+Pj4SEBAgDz/8sPz3f/+3iKCyjS2xV/bqkUeU/s+sutkWYOIdweqiSDDpRgphxSZhQj6WTxOygepXhmJS4bm5eC+OVy9WYKTJNejEHT743Zdugxd839egQwttvcSQKM/NjDJMD1USZTfPwmKzzoIOGrQfx7wzyF4ah6sboAt3ZgOuUSCZhYKBYE/vv/2ds4wf7ywmEwyc5CJlAQjZTvhbJm4cvgaLYkEDLYTFBPPuxSLLG8uaCCLW5oPZDTAAWlNwbncmFuV4Gnfl5VgIwwYw8R02gMl3gGCufVnKTNxtATNuhYmKPQTjuTmUIwew8uwKp0nbHxNgaheOp8crumAkf0e2bRhPpkwRMZvFLwYLfO1e5b2Y5LmQdiMIEKK+R8VbD3MzMRGlJ9I5jI2xqPF+inZLySy0QSj9jCHC8HDbfTy1LsJ6vLJJCckc3YKFMKEIxkoAPRPX4GwKBhw304cWeZp8i1dhQbLacKTsd96ZZj3Ozv6Dbb3fdJOIm5t1cTaF4p0mD+C+9UVkNLMBXQEyQkEKyu4xVL2wBZCoNqGFVOFcnhR4m6lgQqrXZxKrfyfiCGkVMM7y9uOcU93KYujXhnsPV5HVsQTjdz1h1YfDAc1yaIhDOdNjFGOyogJtxNvK5W1C/yNe62XJVZun+Kc//WnU+bxQioh89NFH4unpKadOnZKuri6prsbgffjhh///vIRDHOIQhzjkP0auiTxFDRbV/jebzdLT0yMzZszQFRRnWbhwod29Fu2yT7VyY8SumHEEVkyZRYWWuCYXR+eJaWVOBHTDyB1za9j4Y7RT4/mwJ2lnf11dMiuTDWL9Ya0fbQI8xumLyxerkATDJ1RzaZ8BdjyXc6qOwOdMxJ15gOx+Fb5bbYGHmZVlR+8ffihms1kWrgD0pEFHrCPeEYBz/DIacN9sI1HPuDIxCzOPVCs/tQG/Y+KDrsEYayOoyl5pLicPQEEjT2YpB4SJNfrD+2dUiL15RnIz0oA+TI4A+7UzH55+s7/iQQcVAXbtXbzYpt4zMnpk3DizjvGoMbpSt8CDYtIOe6DWXVZE5Fgo2oAhM2Zj6rxG9ULDW9A/mCyenQU96iDnNHjug0VgWDKKuNpX8YiaY+AzcD/SlU+jgbfrAPSwvAZkoaPz8Q4JMfhtbZPyXNPb4PHvagEknpZmp7/n5Ih5/HjJ7lCYytzGDEe6dZO3SuNy5QGQbjjqwrl/YTXwCjX+Uu4aIsDQJLKvFPMDl1zbuAroRfYe6CZjDbzD1jbAoAHl6j3ZZaVw0bE+9KnYYHo3Zg6RAlavxbjSyG8LQ6nUIymON5ROBnp7WTKmRJsL5Sled9111r8nTJggbm5uo4g2jz32mJhMJnF2dhYXFxcduSb5AppxlF8aG3HofWzEofexEYfer00ZU/j0Bz/4gdx7770iInL77bfLDTfcIIWFheLt7S0VKtnExcVF7r77bvn73/8uH3/8sYigrmlQUJD09fXJfffdJ2VlZXLmzBnrtbVScJckRqOI0Si1SfAa9hKJoX5AiScwfT/dh4K65NpxXJL5I/zbfUtg4R+1IEamxZr9VmBh92M3sBwXKTlApiIB8pXk5SWEwypLqKEcgy3q87KbSt7VwiZYywWzYC0n+1DM6Ai9HL+0qottwbCUM+EU6OT9T8ziOWiWfWmw/DVrs7YOdht7pQ3heD8f+lxC4QVW+iK+F92BmFKlAZZ2fp9yzHG8UybE2JhA0k4xW78O6GC6P7nuS1DNe+Q3cPmy25T4bMYRECBC4uF5NjXBA2JHlh3VvFJ4h7psIwq+BJUrZKCSUMR9pJPaheSJpSfF7DkgrUvgEQTkK3G9vIEs62cHO5CewKSeViPaIJZyXfPKob9UZovY2BtyP2VU6d6JnHL2TlOJ+OQWBzQmPoc8CFHQG25TPs49AM9z/nwcs9clJgSQ2fNldzZ4i+KND4fDO1yehD6XOR2eEcvgAw/JoNksA+ozHdpCHlwhJS9zuhmldXGsjVGopkR4h/WzkOpSYFI80v5ALMqcI73QB55uYwzepd9I3uEcesYthdbDqlD05wBtjmJYiyQ2hcZJA85ZF46xGUMxzW2Z6Jc79nuPfnAiBTXstXnLy5IxhU//93//V2bbyClh+FTz/s5/zLa2Ntm1a5c8/fTT4uXlJadPn9Z9/+KLL8rChQvFltiFT7U8RWJAHWvCRBhrURYxHfPxAJWA4lFFx0vXonMxorfZi1YJwsycvqyQiUae/bv1M94/j0l8XKeTIZFqIyb+yHJAaZxMr81Xk9swofWHYkKzMkhF5Ow//mE9fvM1tIVuM2O6TsIa5TpHkz4fxktJ6RFXV7PkBVLtRnWriNouLAQMgcVGDeIPxh5p0LSvxb39qogVSttQlOxRFreZOTBACtKwCCTH2WH0Eg5d6YU+EF2Idzi1BmxKLeXKeSvafF032HWbk2hSZ8uJmZzUvgFrAZUx/Gb9/gigzd6UFNv9/f33FdYvJ8Fq96COqtWhFdGTXpj8x5dgYhe3WcB+6ENjR3JOKNt9BatgMA7GYLy5NWFyXpmDSZ4XPe2enObGLOnWGLCkdSXcKBahKzFmIYITQejavpGTqaYrD4jeZcts6v2FF5TdSea1KH2hOQX9gO1KhpOn+xBxj6ym5Bh8zjrgxVJDM507cO66nRhXm+eDsLJyD8Y8G6EcUuDCDmzzeBsVaLmgGIYGGxTcT/n5dBDwfhjiG/1hiFuZ8g22yyg676X8xnSmb126jCl8qu2TeL54eHhY8xS1WKKbm5tER8MyzM7Olg8//FBGRkbk9OnT4uTkJE5OTnL33XeLiEhubq7Na4so5ZccsMb/f3HofWzEofexEYfer00ZU0+xr69PmtTI7u233y4JCQny7LPPWnfJmDdPsQRuvvlm6enpkc7OTmvx74CAAPmv//ovm0QdEbGScWyJPU/xySeVCh/rgmGNLCyCNbIvSvU8uK4TWW25paDb63bSoJ0neJ+4kAhYVNs2IYC/OkmBZPt9cT22xNk7Y1o6OxV5W8mytLP/mmYh8jXYggsx2bEs/clyphfV7fYdpcA37Ynw1k0mOx761q1KhQ+O8qsuxLoG/J51mtBNFUfs7OnIzJxT2wEneZfit4NzlPZg6I7hbuZUsecRu4esUYJxdEXRZwFibRxQyAyM/nChFk6zWhgOqzy3BlY5e1XsTDLEqj0K94WBAdt6f+klxWPhHd61PLv6OfBeuNJMfajtWH2YAXlvvG9eZB30XhuBttS4bKx3txbKnWOWGeGn/avgibs3ECGKB4XmkvDGlJxjQp28sg8LVXQE0IfeAZA8zA3w4hu9YJhreg8KBPGksQl+hq+vbb1rlYRWLhntWXG7zY6gMUzPz/nL3A/mxgDV2LYfCJfGYzG3wcs+1IT35kJCkxvgoa8uhIfOfZ+bhuctDZZOXwx9cPiDUQOeQll/uhciwmPjKmXOCfEi5Iaw/MpNQMzId7osGdOYYlVVlXzta1+z/n306FG5/fbbZdGiRfLkk09aP3/33XdFRHREms7OTrt5jiKXl6e4ZpESY5FSfMaT1S6LApd4Ubwj1YiOm96GvL52A7EgCWMIoRyepWnoUIGEzGmjjUlZecsol3APOitDLDzhcvwplfFW6kiGQuV/tyxMNBWhgLhCYtCLN6+izri1Bce0IMXEYDKU+cp9/CqQ15V5+LBNvVfd/oiYTGaJW4XPBvcrhsnmAcDG7aHIYxv0weTslokJ3OyFQdibj3f19rI98NzUAM3AAGCjxu2IsehGMsVBcuOAREQQJD6fJpjkxWD1FWxSJiTetir5CJWpEvrhAIJG6fPRR9q7MXnyHK/beWOWCpuTcbCxsdGm3ufKITGLu7R2Q5caEbm6ATBvZBug57BSGETNsyivbxnavsSE65UsgwESQROhhkKysTfTQkYRLXIbLeif63mTX8YaKdaYcUQxTAwpWGQ37wGUzrhvtBfG76k+jBnvfGLNUh8PmUNTpgand2MlC46AEbN+ve15xmJR31+1vpLJMmgtBUdh5mI8z4EDOJ48gGcODMTnlS1YCHmh03Q9ew363uGdxIXwIoyTGMI8n5Tk0PmEidY2oE9qcCsn5nPfXBmMcbWuCAv75g5MdMxy37cT81aIaVi9Nt5xPVmJ0XWEDEZfw/BpVFSUvPXWW/LWW2+JiMh3vvMda5m3oKAgMXHugejjioODg5KRkSH2ZNq0aXa/c4hDHOIQhzjElly1RJuXX37ZbuFuEcVrHB4eFmdnZ5se41e+8hU5duyYjV9egGjzxhtiNpnkaBe8hgQjXUOzRimXpqwI12HoI9qfoA9m1xCjbHUXLGD2+DT4jmEKDlin50Nn7NWEtKECTvRa5MBVpsDqP7UEhoR3hfpbLiNBUGC9L65hZ0tIWR4OD7Y9GB6sZjVyibveqCiben/11R4xmcw6Vp3mXDMTk6FMduCYpMeImQ6WIU+iNgbBfK3NQlqIsETttb4Qlj8H/tmMbl4FL4S5OGFUFSV5mWLhFuxEv9h3BFb+QtNoSFdExM0CT/FoBazyBH+CGsmTGY5T2sC5GP2iNy7Opt4//FDZT9FcM7oNmWXK6YWsa+9wPL8O9iISkrZDg4h+fGgoCJNDOD8zwwBUYJcRaAF7QOyFcH5fRorq1RDs2j4Hbc7Cj51cgbFRvxhjJqwL+mn0RR8P2aJ4JO2b4KX4tQFq7Q0NtT3PqAXB++crv2dCS8AB3Ld2Fp5neh/NQwxfEWRQbcG8FdlNpSXVwXusBe0V201oCLOnKEaxrgr9kNsmtobCJ9wI2iTBO4yYgJYwusGv4JZEawBX4KcJ0Okm5dk/+gidiIk786g61JUWBB/TRbG0tFQHn2qyaNEi2bt3r/T29sqECRNERMTV1VWmT58uJpNJXnvtNXFycpIzZ86IyWQSo9Eo586dk6GhIes17rjjDqnipHQSe4vtyy8rMRaefEtiAIMeilLg0bnhBCXwPYh2N2hEIjovbozZr8zHANsxQINWhb5St2JC5sT1mT4USyE2XKUATI/upkmehHfe0GIizKDm+BQnRO/ywgKe7QWINcMLC8Uuy+iJZ/ks6CozN9em3p9/XtF7qokYoipKcKgPC/PcYMREyroABc2owgSqM0DIkmgOxHV4cCaHKtfcXIjr8ULMl4vcRPE0Uk6nAZMNpw8w0DF3jrJAV9cAnGHocHUMJr3scjCHeTIK24826F2LNjAXUfkudZI6ZMHEUFlpZ7eGRYvE7OamXwFVi2y5YOLjRZHjunfcAaNj5Bt40NYcTFDcn3hcaYj+Qi+0+bYmpDPwpMkLXvZi9AFOMWDRWNOHNiFUwQsyPxMzaHu3ktHD2xLtxcTOz5WwX1nUdFtoURGIzOFh23p//nkxu7tL7xyFQWwupX7PHY7Z5LTgsURyepStvcpErLH6vAgsuLyg7IpB3Jff+5QBUKX3AduwOa+J2jU5XsgL/ux4GHj78mHgLewmaJsHJ3cYTRdkdWrFE0T0dIQrTd4f05hifHz8BeOCrq7K9kDOzs5y7tw5OX78uC5Fo71dsbrPnj0r69atk2effVYGVKU5dslwiEMc4hCHXKqMqaf4eXLmzBklj8qGODk5SVtbm1x//fU2v7+Qp2gPPrWy8Sj3h1mXGtTCe4GxJcRGXtASeCbt+wFl+HXBet1VDuuPocOZXpX6G4royDq5VfgdW7os3mnwFPbNh+XOXuvKNQrDbtd2sO6WJxHsy64Mu0/sOpCnmpwGy60gR4EOq9tgbQYH24GtX31VzCaTZJfC09UgsPV7wcDduBheZ3Y+Ps8IJ+iEXLVD8fAgGXZjD07jODCzk9sxoQtMZLc0MJG54htDprpdMlhnakdJ3462Y6+DVZpABbizjwAtyDDZttaDMpGzmCqK15i3Ap5nb3i4Tb3fcEOPODubdYQurR+y08Hvtz4NbVDdjTZgj2X5FnhWu4pxDkPzh+MUT3e2EWODi+X7dVOyOA8OHnx0vcom5AJHe6nQMsGCtYmAYKcbCXomHD6vAbrmiAKHCziXcke8CnkzkY0asndgwKbeDx3qEQ8Ps7X9mYU8dwtQguFytKEzEdbYzetfAYjVvYIgU3KdMtQiHByi8V4LVGe1CR4yI1KsduY08fhY1wf0oiBKaVOeNnhcffe76Eh33w3XnR1ClsgWhBSOeiWPup69/an9CNW/HBlTos3nCe+paDAYxNvbW25QN2N1cnISHx8fcXJyksmTJ4/67YU8RUf5pbERh97HRhx6Hxtx6P3alDGFTz9PtLJt48ePl88++0xOnTplrVyjJfVPnTpVOlSigdlslsHBQRkYGLCWgrsUmZvYL2azQTKyYGpkt8A7kKhAERGpHwA2H22E53esA15A91ZYbV5kadcbcM7yJfDQBgV5UVKnNMsuA6rPLA/FuYvJMNUVUaZYH5tRoRReiDbACzlWqlxzeQwRUurg+jotgCU+8jG5xGQ2dq6BLtgx0oINkcVUbkrsyJQpIp6ekhGMMw7VKZbhxijyvLoQrFizhjyQNRTIo6okc/dQWkwE3KHDVfBkNvso3mRzFDyJoCrbVTOOUKyE4ybMU4iLQ1yM41jlNUqfYis7fQ+8gt5ieAIF7B0aKd4SF49nPIC2nmdBTPGgRQ2oFJKZ70Z9i6Q2r07MJpMcbkOfnH1E1QMzasgMPxqH9mAHSUx4sV0ryMtLA3Fi6V687ya1r7QOwDu84XqUZhz5b3iE2b7w+DPaqDOTCxu9H2PlUKKis7kEi+iIKnUtOCZvM47aJnUxdLaZ8hd1exxq8S9SxKFi27pmuaf1JTGPHy8Hu5SY5LxCePrsojrvpLbnQCiRADgcXLCGOhw9k0nlYTEyUV+13Xq8bTHNcYX43dE4jJ9tW2mO4CDvABh4GgKzYxm1Pw2U0FB488yn4YJUvH9o72K0acJe5fPdTfiMU5LZO71ST/Gag095x4zTp09LYGCgdDNLSf3+Bz/4gexkpgiJXfbpSy+J2d1d8gYQqWXW54wKZQHYYQRksdIfE3lzBH7HkAgHozlZmcthMYFBQ0d4kUmOAjx1qAaT+lwfGuyMLRBdtDIfv43uIqhRnbV7ozAx8XMwUjXdBNis1YAFid9z6QFc59gmxSjgcmz24KSe731vFOFj8hwFSu2MwSLTvB0TMudQzbWgDU7Fow28u/HMq3fimbf5j2Y2Ll+GQb8uEwAKzZk6piRDajxfuRfT0s+rhvrAnMPH8A/Dcrq8QxOea1cOnovvuXQJztHyNHny3r/fTn8/dEjMHh66hNjGYkVn3DcZOkv3Rf9hiP5UFXS9dAt0zWsrkyG0ghGsUw5F6Aol9BFpjGbQvDkwBrg/aPYww+Trm1A4YOill6zHLm++iZM4tkArSEaWu62Prc/Lxg8vpr1dXbb1fviwmD085LN77hERke89jCmYYf4f/ICITOAQ6jfoXWwnb5PadPYWhdDHxk9vOYx5cw4MW2745Tk4f5eJ0t9YsTRhJG9VjJ6CeCxsg8uwiOlKUnIHpusNRiCEwuNDa/ajy2CwJh/AYl4QQ+9wgVS9i5FrBj7VRCPfiIg0NTXJyMiI3H777aNqpN5+++12r+sovzQ24tD72IhD72MjDr1fm3JVe4qdnZ0yZcoUK3wqojBRtVJvW7ZskbWMH5L4+fnJRx99ZPM7u56iuq8fwxOnfNGRNZhs4SZU76gvRNCeS12tzsE5nOPHyAPvNM4Bes3jZNr0wuBjNs/tNQErMEcQpMiUbMpjW7kVXuZPf6EYEr97EV1gYSKINnxtrqQWMh/w3mAF4FiuLLPRpFhu7CwlJtrWe05Oj4wfb5aF5Qj+a/R4s+UULkAR+WwTqgdlxCM3jOVQB6xOhlfYC9FUyYgA7yrCtHTOjWTYiq/t3EDQEUf/NayHFMl7TS76GWLgtz30kO1rsCfD1jqd09imeDUhgZ/voWv5odFZ8PjWRymeIJf+Y9LJwhTQ6tn1YwIR6+ngCqQg8RdlJuWeutxeoXYkCKIsHjDejAZKm2AdULikP0XxCnn8zIzB+7f3AX3y2wmiiM41oSrVuaFoJ/bAtbGcMYC+2JyGZ/Xxsd3fT59W8kO1V+RUBV0bs+vMbBRKxt1VgbG4vJRgUMIWq70UBCfyAHlQDEPxICW9MzmJH2tGILzTkETMJxpJh+c7RgK4K3MYgYlezmvxjCWJ0Ls2na2sgsffvgWhGT8T9Cx2yJkXK1f1otjd3W3dQ1FEWRA9PT2lp6dHnJycpLW1VWJiYmwufitXrpSf//znNq9rL0/x8ceVmoTr5xM7jeIWO0qVgb/SFy68rqW549Lo6Q3F5MwQUmQgJvyDR4C3a9Xz+dqHGmwvsiGb0EkyfNFJtA1MRUSP+VCnb9yiQH08gTKs0T8AIMF9LyXs0oRxsJTZpTglcr86oMhoydyx48L5cnwBbVZmhTFFkycwRhSYsUvxjIxuyrOkOPHGcKUtGSblBZIh5JXzqdQdBUVaUzB5MGKfvYbO1+JPNBsMU0kwhs34Nbl7acxSEdG921EBbJ1Qp0BX7SmArXbtst3fX3tNWRR1id42drjOKEQskDe19gvF5JM6B5OSLShTRN9k0xcr795ZDKNqch/l/9L7rd8C+HJjEuXosrL5wbQxy3mqAzDwgrJobNA1dIvlAD0L9a91OVgEfvS0YshMOn7c+tnuKrRpa6ttvVs3d1YNR17cc41gFaensWEDWNbchWdrtMAQ5gVq2yosXFrf4hza9eEE85OxctgHuuGFi21shtYnH6E+qbL9W1fB2A/guD53BrYueIwT/ZVhay3s5FZFzgH1kWaBHoLIN7gcuarhU21R0yDTkZERa5HvkZER8fPzk87OTrnxxhslXzXbNBj1gw8+GIMndohDHOIQh1zLclV7ivfee6+UlNiuzOLk5CSHDh2Sb37zm3LnnXfK22+/LYODg1aizV133WXdqPh8sQefHj+uWM4MFc7wh1V2uEExQWbvB1ss3QhLKTcQMApvVMyOWm8pWbpMu2JXRX2A1HyQMpigw5Y4ezWRpbZLYy23IPBdnwgPIqxD8RAOEwNwdgy816M18F6ZbWlv77ygNkBlAWlKcL+1guFY04X3l6sZXbGFC1zwuzLjk/dj08FhpFNd6b4i6GZHoKIzRuLSuyhoT0SCeiO8gLAWIiwR3F4SCAiYdaY5XjO8QHDgjlYiqDTEMjOYPBZq+OE5ICAx5FSbpjw7w5L2dmvQiGUMPSzdroQLdlPx9eFV0JfzEYzHzVV45nUD5BGw5c+Nw0iKBu+xO0JEMS5uzZdLHUD1GHavF25BmEO75DZfakfCu7lYPt9ep2tiPuXRhr08lgP6VKic8qGb49jTsqP3hgZl31bVdZu+Fb+pLcf51U3wXBlVsjdvrM8ZnVMtAnhyXgXa8WAM2peRCR7bzO5kwhf3Q11puQj1HGJXDa9B33QuAlHuFLG0vdekW4+50D57yq0diqfMqecM0y9sI+b9OpoDLkOuak9RW69dXV3F1dVVPDw8rN+5urrKxx9/LCIif//732VwUFHgl770JRERef/99+1e15E/NDbi0PvYiEPvYyMOvV+bclXnKbq6usr1119vjRmeO4c8pokTJ8qNN9446jeNjUo8sLOzU37605/Kj3/844u+X/CKWWI2GCSErNuCLnhWyRGqJUkWau4SCpKn1VgPpwu8xt5CmJdlVMtzRgyA+jJLLH2ueFdRLbh0kC/u09cHrH26hTxPeu7lRRQDJLcqai3ep6NDeZY9i3Eq5xKx96R5lSIilji8Q9B+vOexRHgL1vqXR4gsYEeSjYfFbHTX11QtV37v5Q8SSFAcLOGVXHKF4l+6a9TBi/cJp9qRxCLRdMypI7UNsG45pshxPwPtaTdAHuzMNniQYfNxTn2pEl/8ZMpt1s8mUa3eolW4xo4seAWVTQiQRIciJuPchphR9Xx4REZ1RId02yYfsbw5Za54eJgln95xd4Tab1LgWTkLvITBeHiHSfTeUmW7/ElnN2Jh3byX6B5Fx4fj8ewtREJbnoJ4bHQO4tKNa+FJsCfO4UXzHsUL6qXi9+YKRpzQj2aGA8lYn0PVk8jV5jSlyfvhYR0OVTyvDgsWu+5C+XzJzRUZN04Wtihjp3YOPJuVmei/O+YjfhaZiPkhPBwe3z6qmZuVZdsDLjEqXlnefHhqCx4EV2Lk2/BUo8nL6xXEKMVOetviPRhXtepYn3kEel8SiHPn7UdH86brDRejbTqImMPefYDqovtSigeTeCSU+t8VylUNn86bN09efvllm9/dcsstkpOTI/eouT625L333pMgG1FXu+zTmhoF1iAyyuZuBL4ff0KJVz7xOFTG0A4zEudunYE/iHLa3AdYiCEuvyWU87VfmViZ9ceEC+YU8ES9vI/gIsaA6cf1PngujQPDE//krejQg5twPV2Am+CbsHKwAev34JxqozKIIxuwMPXOmXNB1m90IuAiTWUBPjAGho0wBpz3A9LS4T80U/b7g5zkPoCFprkbC791Jw2mKjJmS9DovnDog9nAlQZMWNGFBN0Q5r1ylQLKcN4ekxd4NxFdXTVmCdLiX9mNd4uOwqKlkaO45Je93UmamnrE09Osg6SsZdeo8wUVAX7niYgJX/Y4L2FNVHyBXv7QGgVqn9tCCep8QRqDhxOxEDF8aU6CcaYLP6gkjt2l0JGusAQJl6dLXYtFIC8Q7bg7EAvV0lCECFJzlLGUl0KkFcL47e2SoZV5s26UTe+6IxzjaeUc23m2rCaeQxYKjQkiB7XGUXEAVQK6CcYnghvvxMKkVN40QAfl0tzSfkDRjV8hGeQ0QTWaEH5gBjijwRwWKUtB39jcpyyG62iPzHM332w9dn3uOfxwOebsy5Gr2lP885//bPc7T09Pufvuu8VkMkkfg8uqjBs3zuaCKGJ/k2GH/HvFofexEYfex0Ycer825apeFD/99FNxcnISV1dX8fT0lO7ubuv2UN///vfFyclJPD09bS6K2pZTlyRDQyIWiyyvg6XB1pLTK6+IiEgFWcVsOUd3kFXM1WXITA0iU+hwA6CH/fGA3dJURyE9Bjlvm3MA0ejKh5GDszwrCX9QjlN7IDyZUIKcDmWqENsaeiHybnjbokKi5Rt24riILMjZK/D54Sz12uzK1lMOH8unn4o4O0tlDtzuw3WKVbm1GJbrji3wGjnlpIg83T++jny/n2+AR79+Fq4dRNZ8Z5cCOVX1wYM2ReC4gXS9dDss3W1pgK1Xz6EUHnKTBi0I2WteuVsNYE2DCak6w/Hwelxc8Kzf+Q6Qhdw+5KdFk6m9cj+8ix0pirW+sZy8KCEvlKS2VsTDQ2R2KBFMOtQXJk+jeQU8td5ZgO54y6rwcHgjYcsIJSHmxuAskCv2qEj/3EK8E6cAGRMB03oRaYRhwcEKeJ7tvmgbv1LluZZ61eDaoUjDcS/FWNtdBYSGAQces0uryPuPX2w9zNupekwWuK+zF6O9IiNpPiC5x/CGmA0eGGCUqtBN6MHRFhj1jEhtW0b9jRk17A4TDKGh2fsiKC+a0IjGFLwfF6MXQ6D1cLKFHmAVzRekNC3lZnUi9FFCpS+L9uJn7BE2VwAq7+igOtbUx9eq3SR7K7z/DEq5q4zCnI1RdXlyVcOnbm5uujgiy/bt22XJkiViMpnEw8ND+vv7ddtQ3XTTTdLEPYnELnz6/vtiNpulpArwGleNbz+iTOr1ggUqzAsxCd5Xb7IX5f4Rky29A1AMw60JPoAzNKZkQhyu0W9BbMZ9Fiadyq2Ac6LX0GREq7nTL//Hejzyze/gHC1YwkETztKlQeY0DgPxO9+huooUd+RSZVpO1LwOQCC9ixdfcLcG/r02XzC0NzOO4rc8qggq5hqYXILOu4smEmLPzdiuxLkYzjmUT21HBtewF/oFi53qYLr21dA9nsMY4eYm4MR5hrYDKig/lhb2/ggYI9r8yghwRISd/v7442I2GvX4lWrlzdyDRaRkE8UneeLluC6PNYLjMihOxwinpgceX+5GwMDHKrBAxnbR4sIKJgU2Giiu1638z12EF9OVKRiztV0Ys5xGF9mBhbPEgIWT+5TWvpHh1F+IYdu7bNkF5xnN8DhYjLCBvjwfjtkQ5rbdvZYMGk4E5Zxf7XOuTchtx5g+HR+sAJw8bz8MGjZ0jhXj/WLrVOOMlF0giCMnz4Ge2rso1tyNS4c1EBRNL90aPvv8W8vCOYBxnSb1WI9HRkZzTS5Frmr2qclkEmdn/SNqtVC3b99u3U/x008/FRGFfKPlKWpsVFviKL80NuLQ+9iIQ+9jIw69X5tyVXuK99xzj5SVKZ7QpEmTdPDp+PHjpampSa6//npxdnaWkZERnafo4+MjJ0+etHldu55iTo6Yx4+XU3MImiNrU0MKZpgAMbT6ALYJEATtdVaWkSxddg+4EgsfqxZSdTCs9cgays9it4ZNSDub4DUnAfLi3RWs5aLsbA447Asr2rnFdr6cLnmQLMSMncr7syfg7/85HgtfS3svKsPBOWrMpFt5BFYsQ75cJYvT5dwq4F0fMyjeNaufeS4zjOQlsdtGJv1BA6BDtuK1HThERHK9lDawt08fezJMWmFSBXvNXPuZSUepxUrfzVuGd+yNiLCp94YGhWhjK7dTq94koid8bD4Az4/30ttsQr9i/TFBjLuq5rxsbOAdGgpx7TUwajevIWIHjx9q4N0NQEmWpijnb86BZ7+um0qcEeSXfASwGyMV7NFzwXWuSda7XemPjJY4Z3Ku7Vrb/X3WLDG7ulrft7MLhj/3Dy5/xp9zN+TwSW8FwhOxS7AAa/0psglIw/sPPmg9vpG4G9U+gK0ji2muoDAIbyBgLsY1a1WIevoWIvZQOOaUl204mMEHbl6P6xEKefs1ZW7nqjzct6wVtEREthFMfBlyWYviv/71L/nrX/8qnZ2d1jqkmqxfv97Ory5drr/+eqs3eL54enpKV1eXuLu7i5ubm7U2KsuJEyfEl3nbqtgr89ZTUqLsGsAxSjpujVKgAHbh7eQe6xqXJzZGqlh48dXWqtkVpEteMGhCzi4Exs5r2zoLpYSswnXMe4ntp+IzSyMw8e+OQnyKF9mNRgwQXnhYIvswEVsnLKKtZXZ22ta7uigeDMVzaugv644X2H17qd+xMUATZa0R0QW2I7jNNEhPX6MVxxwSZVsgupso/tRHVpcDLtq2BRN7Y4sCF4UIYNyyDrQdX5shtLCtSGxmaK6kApAbdw3tmBdNs9lOf1cn51N7YWBorN+lvviswAKjIzkRkzw/A5cgri7CmD1UBcNqbh9ikNWhysSpm8woQF/bAHiN4eSSLhih3OELshB+0No6oQ/v0BuPd2Cjg1FG3oKLO0FuEeJczBjOHlCherKo6pdgQs7Ls61362bmBuX52qPwbDy3pM4iY4DnJFpRygyU4sV1Ycky0bZgYiMxyAftOGxCO7JuuO9H+4wuYiKij0dXdimfs069++Ao6CwNgsF35FOqFRlgPCkeDlbegY1XnmMX+lAxjdmAuy9HLplo8/zzz8v3v/998fHxEV9fX91mvk5OTl/oojgwMCCurq7WuCIXA3dychI3Nze54447pL6+Xr785S9Lc3OzrFixQrZu3SojIyPy0ksvXVKeokMc4hCHOOT/tlyyp3jDDTfI8uXL5bHHHvt3PZNVFi1aJPv27bP5ncFgkHPnzlnP0RbPH/7wh/KLX/xCRESeeOIJ2cT10VSxC5++9pqYTSY9pZRMp5WbFIuRYQ12RCdXwVppj4C14tdEHhT/mM1OMuNaB5T7BBTCq/v00Uetxx6ffGI9Xr8dEBHDP+xVscXHXogW4I41wcpO3QLYLC8eluchX5Qvm+tLkCJZpBoUKQKjcHUirt0bGGhb76dPi9lslsPFgJFmJype1qk+eAxsaM42kU6ZYEDslloTCCjTW+A1TF4Cy7yzQ/U4GZ/ipCzGXclCP9SNd2WEgA169kg1T5Ah0AwT5XPRc2/bA8udPWW2jNOPAKJqzCIPRxVGHpYssd3f33hDKWvIO7dsDFbHG0GMJQN4V+7vzNY0GymGz+OH2UbsTqqu7NxM9DcmXJQVE6mK3IPs7egPGcXwko5mgmGbUKGiGtTZD3WhL8wNJdIVYZGtK5CHyrmWrB/nLjAlI2cp45T7JbdveLideSYyUswuLlK7RxlH04tshDRE9A3OHctO7mpCEvrN0RjAuFpBi4xZGIu6vRLD0Q/TqwAn54bD6+UQDMPj3Me1zwP8geL4+WNMt+dgDC4twhjcvYJyJglKS+0grztc1RHpZ/lavO+unYQcncdDuVS55F+fPn1avv3tb1/RTS9WAnUcab1oa7lFhVA0b1JbEEVEfvWrX9n8raP80tiIQ+9jIw69j4049H5tyiXDp9/+9relpKRElnG12H+THD9+3Frg+3wZN26ciIhs3rxZXF1dpa2tTV599VXd+faq4diVxkYRd3cZ3I/gsVsdSDXz5yuW4fRiWJSpNQjg5/mAr+3HrAIy2wu2wEo9QBT1g6vghQT4qlYhxUzGD5EOLLDKOZ7ExblL9tgm/bDnoVG+Y5MC8Q75bnQCTMK4bnwsfXAX8spx7dQ6WKexqkHTHwwPU4RSCkje+JuzeHg46zyFw0eU5+AC5W1tlBLRUIhj8iRWHoBHsGMJWaBk0rJ1q+1UX1yMWKCBdOpFBnpIE9p3rhc8+14v/DagCR7LvCx4MlrYs3IvyBCDwbDK3eLx3Ku5jKAXrGV2ZucfgHcYQtV6yuoUHbGDtoPCyCweHgpJZOMKeD/1XQpRJywUlvdMKgIuXdB1WTfer6oK/WY1uc71A4g/xR1B++eoDjg7+e5F+H7zdhBw2MHkbawa96KqUEIDvJDdPko/XNoAYlZTN34ndeT6ka4DqpAOsH8/2tR5FsgnvfnQhRb/cyvEc3ebiDhkR7bf/1cxGs2yRotj+yzG9QXej9mEzsfcgZQmtE2QAR710VCq5NKNw4w2lThYgetl7qTUh1Iv62GuP4W/0uCVBQn6iBjIVdyEFA6tAk1AA9gwaWnQXfQmeIfcptuOwGtdTd59Fs1VkqVCLYTW7OrbjuMcIIpXWNDm0uHTp59+WrZt2yb33Xef3HrrrdZtnTRZuXKlnV9eumzevFmeeOIJm985OTnJ8PCw/POf/5S77rrLmpahxR2dnJxkw4YN8uSTT476rT34NDGxR1xdzXK4aHTZLBER9xUq6YFwyn4fLAruXRRUptVqmBxyRuMKGmA5HtuDyVL7KcMUIV7olJv3IPDPpJDoHJAydkVhQmDIi4kSWn7UPH9MLnktlKRPJhMTVXjC5Vwqhma19ZzfwWCwAyfdeaeYDQY97KahBHSBViMmhgATkRDoIUpC0f9mmvBe64rwXmyvaDAZvyu/nz3omUur6YQhL87XKlIWDSZDzRvAQA7IBOOZ1ZDahRWtNQnvxhO40wModTjytrq3KDV6r9F4wfJ6uw+MJu3M9qKyfpycR1ZVezwgXGYFLhwA7B69B0ZRZSGR5rQ2o+RFnYHVQpAiG+D8LDTG2oMB8WqsXh5r0zuwmO1owETNRuJMAwya3BYs+Lxws7GptW9rFwpMMAO918vrgnqvbVH0ziQvZi/ryqlxv6IOGm2A0V6ZT89GFzpqUhajBBPCHq2+IKEF5BD7mPoN78nJY4aNVx7/GsLL78PEodyteJ/KJhi40cH4PGMLPs9eZYNkyZRtntjYYrSzs9LFyiV7irt37xaTySSvvfaavPbaa7rvnJycvtBFce3ataMWRU9PTzlz5oyMjIzIp59+Kt/85jfFYrHId77zHXnhhRdk/PjxMjg4KOfOnZO3337b5nUd5ZfGRhx6Hxtx6H1sxKH3a1MueVG80JZMX7QcPDgabjtz5oyIKESbN954w/o8GiFnwoQJ0t7eLk5OTmJk8+Yi5PrrRdzcREfxt/gA/rEG5YkUw0bsvhWw5pZugtVr3XlARAoLybffjh/H5hM1XTVxh2Pg3SSnwDssCAfEsa4QqRfRRBoIJIOavSBmR8xTE302B8Or5HwuXUHwtXi+uDhAeuxJMadJo9F7ryX4dDKVcCIpWH5E2U+R8j+tD01wSUAcp9fQS5HFyAb1bIJSD6cRGYXd3jmqW0aWZmIirFX27KLrKFd0D7EriIPengb9dVMuFkPbmpSUwzvklAxd2gy9EBcemT8f8N7IEHmtNerm2nb2EtXJK6+IjB8vS/klTaqZvwU3q84CNMmkIkL3ZGEgiE/7Wmx7hwfLQb33jVH0FOGFa/C1I7fCe9lLKUDTmQDUh/b1oXbfuIQ8Jk3K0RhLlsBTdK8iwhb1Kc7/45y6QV+Ma62LBnQQ8YxzYaJtFxx74RWzjB9vtiIPn7qAwR+pIl4iItJNMAuzfeh4O7137hE8W04OjrW+1xKI50k3IrRQnwavPKwG48TPSJ4qkQK9qaHWL+m2HmdsV9qXERX21nftxbhaPotSLyxI9GSCEw++w3XK+wxEYL5LrsOxjll4hXJFyfvaTzkt44uUadOmSUdHh5VMw3LTTTfJ2rVr5ZFHHrH52wkTJkhiYqLkc2dS5XN3yeC8oz50Lm3N2ZaJznK4Ag09uwWLX+scLH6cH8THDC3OrcAkkBusdNL0OlooibVa0gbYdeZeQFid29GhJwcC0tFNkIx3qD22nmpwhu2kRZsTjhinZZ1ScEC3i0WDAgebY/CsbW2fs9ktwyGaYUL4VrUgRy0yGNfhBa03CQsNT7LN+VhwB8NxHRvplHaJfkwM5ARqt0Qqr0eZ3IOFYCO7FSsLC+fLsc3mFg5oeLgBcWdnih02RqGtOd+R8b2yeMVgmlGOia53xQqbet+5s0fGjzfr8k61Z2KUiiurxTYB8m2Nh65ZAo7AeKiNAqTPUKXW97n0GxOz2bjhbsHQXMgRjDfezUYzyMaNwwAbeQVtYV4Mg6J3Dxne1HaHLGCPs0HDDO/NRWosjBTIC4y9YhXWogkm5bsdewFfc1swOZ0XGt6rmXU2meJ+nQIDVDO4+BrOObZ1t64DSF9eDIwz3ebOFBbIbkB/zvBRzjnsizbnPEYu/PDEEzDkRt4mi5A6XlgpnlEzUryrCBrlAckDNW80G/tS5LK4q/v27ZNbb71Vxo8fL+PHj5fp06fLSy+9dEUPYkv6+/vlW9/6lnh4eIi3t7cYaLH61re+JXPsZZGLSE9Pj3xCqQssjvJLYyMOvY+NOPQ+NuLQ+7Upl+wpbtu2TZ588klZsWKFfPWrXxURkfLycnnuuedk06ZN8t///d9f2MMtXrxYXnzxRZvfeXl5yenTpyU0NFTefffdUd9PmjRJFixYID+nSuqa2PMUt25VLGd7VRPWD6jeHLFJT5UChmAOQGoKGKKca8cVHrRSaCK2LcSlBrLOyFxNXQMYylqpX0Tv7jD+w2UqyGt0mqAYDSPvoQscqoM1N1cAm9UGwiKcngVLW+c1smmrkSAo4G+v3FhdnWI529o/kvdTbGyDJxoyQMxSdh+4/AnVgZqcAm+uM5PgSdUNWl8I71FH0KB9DkssIF/w5rRctYXVzo+lOd2x3fBYkvfAGynIhCcblIJnYUiaCR/sBW+Mg/Ucm6VAg8fWcjWXeNse+oIFYnZz08HJ/ZsUaJwRDXuVBHnvyMbF8JAYnZ5sIbIEKad/jsLSZHIas6TZG2I97p6P9hikza656914m4JcffhP9GuGBXX1wZgJwi4hsUn2FRHrmWTBIuU+TsQM5zzF0FDb88wLLygVbeb5KtAt5/fG9pEnxDRpRm2YnJSG/sR6mu4DvTf2Kf2TdcTePyNW7JXz5tI6pIhyCTfvR5utS1Pastliux13eCHsc2oVoE/vOoKw6SX6jdC71kyzfSnEwpMzK/4KC8hc8qJ44403yoYNG2ThQj108uKLL0pWVtYXGnM8c+aMtQD4+eLh4SF9fX12S8E5OTlJXl6ezGd8RhV7Zd5OnuwRs9ms03VACzWYulrqEoHnoOOsXgPHe1sWwXs8k/DCRQOvMxHQmDaRlDVhsuWJZno47tncgnvyelCwGJOivWRfjTLILDO/coKTGLbmCYNnTAq4NIdikg/yV4yCsgoYBCUltvX+6qtKEnl0HyY86yRAwbajRlyfB7VbHwyDxi4MpJDtgIViawDF8DxorYNKjb66ArT6bR1Ux5Hwv8pZGHhc+5LnMVu7+vh12U4T4d0apofCoNq2k1IdurAIza7BIsTx0v4k5Xl5AW1utq3306eV/s6Totbk9hLSeZNX3ct+XtUCER3U3nlAaesJU6i+5ZuYihhqNWdRGIE6+eEa6Gx2PIynmUmK8VRSjHHS2oZxwkzZ9AH0i1Pz0V+826idaEE6XArDTOsy/LocUpw717berfCpqPMW1xsOhFHq3I1+vTwT/XpXHC3wvBLaidNUBir9mafC5gN2mKhGSr3gNqW5oLoPUD+3k7tFmfPCYjBn1xdSfyELb53/Plsfi/MqImpygFGNJb/j4mL96MSr6C+cmhaCx7ssuWT49MSJExIbGzvq89jYWDlx4sSVPc154unpKQ8++KBMnTpVvvWtb4kLKeSb3/ymiIicPHlSXF1dxcPDQ+68805xd1c67eTJkyWJLUKHOMQhDnGIQz5HLtlTDA8Pl9TUVFm3bp3u802bNsnvfvc7+cc//vGFPmBAQIB8+OGHoz5funSpfPe735W77rpLbr75Zjl9+rR0dipWjouLi6SkpMgBNpdJ7BJtHnxQgZM4UY3NaM3FIM8rOR+eBMNMeV5EWOGAMHmKw7SRqkZMERGptSixh+l1ZBESvsoFfNk73NZF3jtXHmfvlDCUev+Z53+kE7Z669fCspuxB/dhFfvtHZ1bdkpg4drLUzxxQvFYmOVnLcvGOKEd6TQCrpncR6w2woKS02DhM1RpbVPyBvL6ABVzUeb2AbyLXx1grtw2tCP3AfYgNc+Lu1bYEYJxqY8cLYWtyqhFaiJZ8VxyjpGLGOWc+i4QLezuTvLqq0pZQ7qJthGwWzm89lMRgCkZ4k6PIm+KYwe8Q0saPOqgGkoYV5/fvAlsXd59wd5+mazgyjZ4itFNNFY0GIESPquXwCNkTzEjiwhpzJ5iZIRhCX5PjQlLIYLhcCSi9/XZ0fvJk2I2m+VouYICJARSn6VraXt9ioiUbUefbfYCvB5UjPfiOYfDN1aKPPUxJpvxGGY/wlwKtEnLdRQRSWihsA65n04eCvN55HHil1AsIq8B9+QSbiXhQAJm7sR9ODSz74Ciq4WhxPRlrJdRrSDKGLgMueSUjA0bNsiDDz4oZWVl1pjiG2+8Ia+++qrNFIorlePHj8urr74qDz30kO7zhx56SBobFdf8/Jji0NCQzSo4mjjyh8ZGHHofG3HofWzEofdrUy55UXzggQeksrJStm3bJoWq2XjLLbdIZWWl3H777V/088mdd94pH3zwwajPDx48aI0XalVseBeNy8o0+drXRMaPl40VsPzXWyhoq+bCTE+DNVgbDGuufStZwgI3q76bYoPhYKD50T54HHcsVQPSRyzwQldTwWWmU28z4HcFSbhe8nZYYjrmEHmNYSleIiLSTTEF9tbqyzlWB+9wZwTO8bNQThgHLVT30/sickXLy0Xc3fWeldVDpA+PNcBT4xhCNzkSDd2wEmc0gISwfz/ikbIEOXhLTYqHsTYJX6fmIz9TjgRaD/04eESBtnRfLiBOriAp099fafewKrRRWQTiJzOovRLIa22kbYXW74T3t3EJvLEu9vTVOJDFYjsnlKU/LEoMZrNkkqewLVD1/sgK5/SMdWsRp1u6DONgdwfBCpT2E9QFL2hmDsaKllbGDtnwAXinXHh70B9t6iYYB9GFGAf9i+EluXeQ56VKZA5yJyM5QE9xq9QijLftxCs5kgkUx0CxVst+RRephOg4c0kaO/LmW27i4eFGOb7k8RChJakFH++ugpfFjx9Eug7cSugUvVf1CsWzizyCfl3YhOuxw8Wx5Jk0JyWkEaHPiEBiWRU87TffvEVERA5SNaJ4GqepJsQXswvhHQZT6FK39RttH6ZVSSpoQzsmU7WkfXGoorTwyhzFS4NPz507J9/73vfkySeflBtvvPHK7nyRMm3aNGlj+E+V+++/X5566imJjIy08SuROXPmyB//+Eeb39mFT19+WcmX4/txIpCGuxCcUtuETsEcFl5c0v0Bta0rxYLLEzsHrLXbbE4k4gmPBCbuEL2rejtIQZFV6CRHqf6oLvlaXW94IPCmupWCxZLJKRmrCJoheITJPcnlaqfnTVn9/S+4O8nhDrTl7Db1+QlGYzYsx+B5MY0OJYITLVzrSwHNbUyj4L+2aDOpgIVe/JA/Jh1uO67jyFCQ7iHVDjEcj+dw7iCCGC+4LGzQULvzhq3eXsQSVGEkp3HYYLunx9O23jdsUDZ35sRnFbIqC8e7zgjHhJi9B4ZJRjxBWaQ/3r0i258gYq5fpzFRGU9m2i83Kg8OFoZViXCSW6EYIBwWYDIHj1MdO5Z0fawJRkVsIkGsdNGFmQpsvy8Li3BZG9olIsL2PPPyywr7dLaPor9OSqrnbsC8mck+aONjFYDX+b04h5LJotoQ4vee24J2qY6DcaYrLUfvyuUfmXjF93c3Ks/IZS15HgzZA6hcF/dh6JPuyf1IG28rl4BQxTA97xa0kXL6L0cuiWjj6uoqr7zyypXd8RKlurpafvvb3476/NFHHxWDwSAuLi4yYcIEmTp1qvzlL3+xfn+htd6RPzQ24tD72IhD72MjDr1fm3LJRJtFixZJRETEF5qPeCEJDAy0CZ9+//vfl1mzZsmCBQvks88+E1dXVxkgyyE/P18eeOABm9e05ylqhA92xML2rB59AWagkLXaaECnZyuVHc/lUbCum31gIbIjqLHYZ3RTWgVZy7nlVDGfjGvmBDGKw4WOD3bB+po3S9FBvwHEHXchS4zNQArSMxWeHRm2TjXi0L4q6CQpybbetVQYtybAVFYFsktGuq70Baxoz4tNXQz4Jc9EpceWwIvWhMkX6wrRLse2o71iDxBd3A55alsN9Lt6GXRZ3aB4G5FGekf2dOh4dx9gvKWzCJ7mTkId7FQwnlfjgaTmU9WWvXtt6r2pSUkN4Eoo1r08y4m4Qh7vcAfOdW7Ds3H5M95ZptUH3j9f89Qs5R29C0Ha+PS737Uee/zpT7g/YayVVbDj7XlV0UdU+I5g6KM+SLNJ2EkhD9otwq/K9njTdTAKEdS3KeMmrALvkNGAai6Zmbb7u5YKozUhj1VOR9SlKJE3tbsFaNPSNgrvUP+on4V5S4NHOc0krw7QN89VPLVxWUGGuXk+m7uXcpZVPHh2KTxCnp84/WV9B5AIJvo0rrKdIqNxoDLmo8+lrkWfy0sk8k862uBy5JJjil/60pdk48aN8sYbb8gdd9whHh4euu+/yILgIgppxpa0t7dLQUGB9Kmwzfnn/eIXv7C7KNoLgO/apaBp6wzA3ncFI9ajQQHzCRLo62N2JT6fG0yTH9fspN9yrIbDcTMGFLhVYwKK6Cea9BQMtPRVWNB4nuZQ3j6q+M+Tf1+f8lveWLR+BTplGOExJaVYYHgHhc5gwCrOTYAlw1KUxZChWXt6r69X0LI2Kl83N1idBAhym92NSfXwTkBW0StIv6VIEB8YwOQxvB8LYfReGkAqFB6dBoXNLsdCeLiYJh2aGQYDYZhY/HG82kLJ11vx8pEaXNSCr1k5+/zB5tZtI8oWGie/r8L53pQAnxqqrA65czDZN2fbzpc7ckRk/HiRjg5AhRlLVPjMDlzvTDHxozvRxxO6MeHq2JFUfEFXP3OO0oeXV2ECS3kVx7rcSSKiMgS40B+LRhCxQqNLlfHLfS+B899IwTwGN3rR4OTcXmacUvwhNEo1RpqwOvM4ttffV61SaixrNTXC8oH3bWxCf8uhcZsbiHP8o9CvO5PwOW+e7ZOGRVFb3DYbwOZsECyK3N8Y7lwXij50KhCLH8Onh5fgHG3OOUybGbPlUtaEeSYpBqcwuVeK0Ec8JiGHNfQPiu/GuxIxWVjyL3kpsyuXfKVf//rX4uXlJcePH5fjx4/rvvuid8kQUeBTjX3q4uJiXfweffRRZbf2w4elq6tLnJ2drd85OTlJDvd2hzjEIQ5xiEMuQq6oIPj/D/noo48kIyNDfvvb31rjhGazWbq7u6W3t1euu+46iY2NldraWjl9+rT1dxs2bJD1dsr92INPv/e9HnFzM8sOf3iKvcsABZh91YA74w3kybSuhZcVUEgEAyYy8GJNwWYOTjtrm7qStc7B+MleRHShxLGFRYCIGLbgoiKMBGmwBZeeq+yAJcaWNpehCxJ4aYcbQCxg43pyt+o1ElGmd8kSm3pfsEDROyNWGYmKZ8xeB8M2MyKIUEMkj2FfQLtc+5zhHy49tXm+YtXyRqfsjbDHkm2Ad7bDFwy7laHwDhsDYcWHVBC7WMP36EE2F8LrYqtcRzzgfD0m7jDOxpiXSlCpTtlMp9ru715ePeLkZJZTRyhvU+1Ph6LggTA7MXo/jN6CePRxzm9zToFXEd2G92LO2tJwBW3YUQWkYeViPOPCFUBA9u3E5zPm4POyLUAseEcZbXhyHqiO6RsKCJgJNdzus+Oof9HLsXeqITPsaU1eA5Z2786dtglODQ1i9vSUU0alr3oPENmHcOCybvRJRn64siLzVSZXAP4tMcJrnBmneuuM09LLriwGM5u7Fbe7VpJORKQ3AmXpGLYOOqD2OWbz8QWZPEWDszIec2z0AO7D11mZqbT7jrWjq5eJiN7FnTHD9jkXKVf1onj69Gm5/fbbJSQkREeiWbp0qfzqV7+SV199VRITEyU6Olq6urqkuRmTdUpKirz88ss2r2uvzJsWU3TfgsW0dw0mBy0fnidvdv3dl1BJMPLtOUE4ew0GZMZWDEhGyQ6uUgc7F9LkEUt0x8PlmCR4Ak8+QvRsWlw3GjFZagsdD7iwYkrl4JmOVyQ7MRYd5KatuMQKzDx61Kbee44fF7PJpIMk3Y6o6RQEXVUvxiQXmUK8a4J5D/tgUuI6ibsqsLgu98fk0RqhTB5cY5UT/Rmi4Zgpx7NYTcvjCTanRi0xKBMPzxeTa7CYHjNhMeU1kVW9vg5GD6/suiT2FjVXmLD8zOpqm3pfubJHxo0zy09/+o71s7ffVmj1XGpOJ/QCvDF27H70t8rFMA6jw6HX1CXQqzZulvti0UwvwmLKjN6QPrRj8ia0I4/DzYuJUazFCNgS4gHGY4lmdWZps4FnriBInG+qWo27u9EuevjU9jzT8/zzYnZ3l9pwZb6YbqA+Q2N+eAmex7mOIElahbmQBxOo+Tk0XfI7ubfgnu1elCZGZR6HU/BePBWxCnjnjfo1uaM/ZKPOTvjBrRtzYuQszInVBNn271f6iftaQiKZbssB0M1UROQy5JLh0/TPCWLm5uZe8PtLkWeeeUamTJki1113ne5zrXzbq6++KiJK8r4WW/Tw8JBPP/1UutiEcYhDHOIQhzjkIuSSPcVvfetbur/PnTsndXV10t3dLQkJCVJQUGDnl5cuN998s5w4cULc3Nx020B96Utfkry8PPntb38r27Zts/v77Oxs+fGPfzzqc3vw6YcfKp6iOQ6wRYIPLLSj8arXSHDoriLAjctz8DsdpYohUzaByYrqTIJVOLlBhRAIvzhYA8tKY42KiJ7dw3lgvPsvuYJlBiqlpcoMA2CoMgtgqBktgP8it8MD49uwd8pk1cq9iiXaHwgr1GKxrfekpB5xdTXrmLG5aQpMlZ5kuygyI4kh++HZZwzAs88eIKvSDqSj7VAQG4M8sM4uQNmTs+ABDW6HB2TPGOYyaGy5r4xSdFxrgn6ZWMsQN9tzyRWU20WYeK0RcDp7AIN9qodH/aw3K8um3p95pkeMRrMO/pvbpRi1uQLjNz2ekuH5pSiMUOIFr2JmBCz/mWmw/EvS0J8OGpX+NG8OEXGIwbH8gG2Ijutfrw+lClr0LNnG9aPOnb2HWJLcgel9dGXIjDSPsZdJLphWxJ3HACMIPj62+/v27cpuPNrvOFWTSSwzV1EKhx3os98E/bofgEPCBKZdBnUccNyAjtd1gZSzuRT9c1ca5oXlKWjTsgbckx9r5SzVW+e4C3cu/pznJxqPXOgkrAXFN1YfUZAWRmV4fioLxPx0hejppXuKv//970d9Njw8LN///vflpptuurKnOU/ee+89GR4elvHjx1s/c3Fxkfvvv18mTpwovgx625BLZZ865N8rDr2PjTj0Pjbi0Pu1KV9YTPHdd9+V+Pj4L3SnDCcnJ3F3d5dz587JuXPnRETE3d1dbrrpJqmtrZWenh7x8vISg8EgFg7EiIjRaJTPPvvM5nXtxRS1ncjZCwjqoqodqsnKWxixo8bVYNgqSq2Ax5KXQhYouxVMllA9u2F/eKEsXDycvcleAzwp8zLEN/PmIJWBd9DunKNYkxy4Z8NudgSC2rxnIFeW0ZnjFHdsD1bMNb94eLiZ8+bZjrFohanZ/VOv27sVqRTm7VSqggP4FKvgPdg4dMSs+vRwaic1WLJ5ANYyp+RsHICntn4ZrGVrkWURvYvMXgU9o7bdTmQFvM3BJfBC3aKAMgRQegPzB5ivlbcGcbbYFfBwjqUoyMlBf7xPba3t/v7GG8qWXTx0NIeaCRwMbrAnw14Ce70Layi3l0z7vDaY8Kl1KmmJdLTPAO+G41bRUfDi6xvgxfNz81DSYr+xR9BflncATeDUpTBfIBF5xeg77MgE+cKbrW8ZHW/mc/mZ6uvtxBR37hTz+PEIrLL3TW4nowq6uCN1bN05q4C0HNsEQpDWfmUDQBdKNmEMzBSKmRJJQsut5WuI6Pca8GvDddYXKdffmIL+ywXSnbdQrI/uk12BfmEvDDx3k3rtOXQ/LsHJJJ4MQlcuQ76wRfHw4cOyaNEiOXny5OeffJHi5KTkqWhxQv68srJSoqKiJDw8XP75z3+O+q27u7vuNyx2y7ydOCFms1kOHUFnmFtHDanOSrnFWCDSKwB76uAJnpGZrcqrDs0kefnIA9Q6hncxFrPmGCxyQX0UdCcpaEIHTI7HYOedKryNo/PGuLyavSTdgA50Rt3+axaC1ohoo9UpZaLK9u229a5NzmxgWDeFJX21B2IC4AHD67K5hXRDsPEpHyzOtmyR9mL8rqQDemQosLoNsBEbEpv9sdD5bcJCxzyApXHqpEY4aYEBkGNyExbig4EY1PNa8LluVSQjKrYOhoPGMXEfQPv3Ggy2+/sbb4jZZJJDLXjfufvVZ2IDhfq1uRiQJeuAyVpsrzBRiIGdBEPZqJO5KATDkNGCvrf6APreT3+GPLb33sU0FtKhXpvJFxTOKCjG+OZdbk7txHjj0nkz4rEQM0xn7Xi8YtCG3r1pabb1/pOfKOX11IINZXPQxjMsWMx4E2W3IhjT+/oABS+kvEbdQGDLXusUPGho9ea9QaN9qVgErX6zK9bb+lhnBGjDn+eQslLoMTkFemQ2O8P/vDkyzydz5yttxqzo3lV4JkbEr5Bnc+nw6erVq3V/j4yMyIkTJ+RPf/qTLFq06Mqexo5oi9szzzwj27dvlxMnTshf//pXiYqKEmdnZ3F2dhYnJyfrvoqffPKJPProo3avt23bNtmwYcO/5VkdYl8ceh8bceh9bMSh92tTLtlT/NrXvqb729nZWa677jpJSEiQ9PR0MRgueZ21KwaDQYaGhqwL3rlz56y5irfeeqv87W9/E5PJJL6+vjI4OCgWi8VqmYWGhso777xj87p2PcVZs8Ts6ir7UgAPciWujBrFqhzeD4uSDTJGQGf6E9xBkAh7ZQw/MVylpfAwJMVq1dHFCSbtzcFzMSLj50P0en5I1cJlyNFeFsi8JFxj2ACvlp3gMB94VYcqFK+KCw/3Ll58YYJTH6zEbQcUr4H1y1Ype7H8nFydhy34HYWAotmT0Y4ZfWEPjwx/XRvtSEQf2dWGnDAmArCXZAsp4+dwWwtjszoN5DF7+ZUMg9dGAXacvkKBolr3I9/Ly8tOf//wQzGbzToiWHqD4qXmrqH+S99zWgwTYGZuAgSWPQf3zuiC15sXAY9I689MtFm/BR7cxja80+QigvzXkOdMypxdBQ9C82Ddd+Lc1vl4Ds4h5rxLdixn9IHkwQ01YwkQh7JVBaOegwdtb1SUTb3X1Snl9QJKFaLIcBq8T2cLjVVyf1Z3wyvaFkMEI8KzWTcMEWv9hvs1O7dBBniHh+swTmZ3QO+DaWgPtxakvxzrgj5iixRI/HAcXDWe4xgpSF8L1MXuSdTvMlqUcZ0dTCUaiaHUa0HfMSNL5bLkqs5TdHFxsW4FxeLk5CQTJ06UiooKCQkJsfFLER8fH7tQrr2YYk9NjZg9PaWkCTlwM/di0SlZrCw6M+Op41KDlnQD2pm5DNfI2wSIkRc3nhCO1aBRY00KlDcYCliLw1YbLUgir52PDsjJ6jzJM7uNF/lYgwpLMaTLmBjPyIyPcaCLVo2SJRis1s1CCSfJbG21rfecHCXGQnBJwArl9zxeOCe0vgq6W74GumMEabkXjIT6CLRj2BZMQqe2KxMTL76c48XCYUTvJkB66Tm2maB6JqLyP6sxNoKgbKKw7ugArMqIqTmFytYVIw7E5fU0Ybh42zY7/f3998VsNuuMIu35eIHI2ILvW2sAywYswfPsSsLzcK7msW4wKGMDCRpTOygn6TMsZy1gISLt4biPzsBjy4gGljaxz95LeZ2M1/Hg4GM7O0RvLoYha4s0yeRyZqK6u9vWe0ODsijyMNKESdLMSOY80FObANfzcF26F+EFDsruilIWt+V9WDTr58BICDtAECxNFuYo9CFmWDPEq4tFqAva4BrMT7p6xrSAH/XHGEw4gvNnlGM+4zlPmxqme2EBL6jCAp5sIUNhHrX7ZYjz558yWiwWixw5ckR+9atfyZkzZ0REqUXaZ2/7ncsULaYoIhISEiLLli0TZ2dnGRkZEU9PT2vdVTc3N5k3b564ubkplq+IdHV1SQevAA5xiEMc4hCHfI5csqf4wQcfyKxZs6S1tVXOnj0rjY2NEhQUJI8++qicPXv2C605esMNN0hra6vN7/z9/eW9994Td3d3mTRpkpw8eVJGRkasi+bIyMgl5ym+9JKyzxk7TrpdDVRzaXgNrCyGyBL2wBv5lLa7evc4VMyeDMe9reQAETkVrkBR3ksot4rdFK7qzfgluRUZFfgt/5RThbT7r5sPT7ZZ4OEGlVIhBjJr88Jhzdlj3oWYFK+gsQ/kCV9fOzDe6tViHjdOp5zGKEWXDFke2kPsT6KT1kYstPWxDspk/W4sBdS3PlNBInbshH346KOF1uORP8ML1XksO2HdpndAH1xej+Hv6aUKZNc5H3Dd5J2w0CvnAP5j49tWWT4RPbGAYVWtElBuBxjSKSkX3q3BVrWUQSM8OPbWo4PhKZZUwYO89164SH/+MzwuJirxhap9FF3y5r86zI/mEadfZlmPf/ITwG48DBjU0DxO9jy53NzgHIwNHj7s5UUH0nPTSdUm9J1Ii4IWbCwGUsBoRlSUnf7+0kvKvq3aAOSqOonoS1wgih3akBZ40SuL0Cd3zMHn9f74PKxbyTc8FQpPUldajjuWnZxMHWOO0CEd5B2qIDOHvTAPsk7ZZ+LqUH5dtkluzHRe163OuexK80TNqMEVeoqXvCgmJSWJp6en/PrXv5ZJkybJ22+/LUFBQVJaWiqPPPKI/Otf/7qiB2JJTU217qXo7Owsw8PD4uTkJCMjI2I2m6Wnp0fuuusuqa+vlwkTJsgnn3wiBQUFMnu2MiFs3bpVfvSjH4267ueVeWP9hmyhCj5qI7QuxmQW4As45/ARxNq0DURFRD/yGBOhYNhqE7DybWuUDttpwIIy2UgJ+3w9Womqjej03Hecl9A7cExRw1t5MqLOv9mCxX9dBGIs1b6YcCO9iH1KWE5Cccb5l5PcXNt6F/lYRMwy8hwW4eZZClzEgypsEwYbzxLNa6G7oCail7Oe6DisFPCTtogG7MdkxDtWMOrGTWdvT+CwBkq5sVFEtTpmudgShqfW90Hvs+sAedkrDMCirTv8rJ2dF07JYDtW6x5+a6DrghTA0DxRJ1RRfI9jQfYwQLYStAmNdUSzYGU4+mx0t51FIGJ0epEIhgQvigxrs/HGpQAbTYBJ+VHd9yAGeSgQRs3cOMVA0G283I2+k+nsbBu21tin2krO/ZT1yBYsj1FufGq8sjiCIWMwLzW2KPNSiBDMzkFFWqFKgtE/Z0bBANLFfezwRoaNygLp3I3ftdIuQjxX6nBfitku3wsDg+fhQwfUUAMPFNYbxz+uMHv/kuHT119/XTIzM8XNzU33eWBgoHz00UdX9DDnC+/ZqG0orMnZs2dFRCHU9PX1ycmTJ2VgYMC6IIqIePEIdohDHOIQhzjkc+SSPcWJEyfKG2+8IWFhYeLp6Wn1FMvLy+WBBx6Qjz/++At7uNOnT4u3t7fN77StoubNm2e38Lefn5/Nhdou+3TDBsWCY2YKm5sqfSshDhYPB4Nj/QnqZQuZsJ2Vq2CHMPTIQXcNMtQVvyVo0bwGkBN7DwlReKdeAfxlHgAU1G4B/GRNvKWgfFkNfsdM1HX7QZjYtxaQ8sJS8kJJb5UmJceKLfS+Pjt6z8gQ87hx0r+WkvNVca9Bqal1RfCEWXf2ChtFF9P1yLXbYYE1rBmY04NtJ2izR8hODRNRGMr07oI1riXsi8DAXT8f3/Nm0Uyoca4CypCRD8s5O4rIBLybQjjeR2uzdSu4L4htvb/wggLjkZVt3WTYC+c3d6FPBGVRnh5te8Fl2XZlEfRIXs2xDkDzsfkK23abP5i2doaMzIuhccWeAuHj0UlAVbT2SNiEPD9+1uw2eMEZ/vCCdY3NngfDHRz/0DgL7NFRZ+gdGLCp9wcfVHaF2ReverrkeQ3Oh37dkmDgM4vuzKRJ1mNPtf6ziEh2Fd6XdyTRIHgeiwEGwKfNA9AdO1+MpPJvo9OI3Ejxgs5VCtoyZQpY/yNDN+Nc4njoioF0o1ybLj+WB4Wq1x05cMa4K5Tspz43mZitlyGX7CnOnDlTtlMQw8nJSfr6+uSpp57SeWlfhDzzzDPW43Hjxomrq6u1OLhGwtm2bZsEBICFxN5hCgd4SLKzs2XatGnWf2FhYTbPc8gXKw69j4049D424tD7tSmXnFT47LPPyr333ithYWEyMDAgqamp8q9//Ut8fHys8b8vSg4dQi6YBpdqaRZaPuTAwIB8/PHHYjQaZWBgQAwGgzWn8ZLh3FmzREwmWZoFK2b3HFiG6WmKh5i1GD+J9YHl73QDbIx334WB0FeD83csQVB55hoElUuSEOeSKvWe5Hklb4HHwGFBXWyLrFTLGirrQG6obwxZUQ2qFU9sjhnkAs2g6w2zd7gJHtiprYjr6IgKA2qssUlffs+mxMWJeHiIO+0gP2+NYugcrCOTNwnPwO/NnmJAE9HF2aonj2VlEsVn1cDFthxMWFxiLKwc8crJZH7vbojHSUZi95DOIilWExmnPGR1H6z5dCO8lINUWWVeImIs2SvIS7LAXHe6Cf3o3XdxyrolqsXc9/l639qcLEajWbwo33B5otKf12+FN8D5lC2LYdUnUNwqk84pqUEfa6MqQOkG8ghUOGQ1kSxkBSmeAkrLt8DD3DWL8gBNGKeVBxDbnrtKOT+BYJR+A7zdjAbEEaXDy3oYuRbxyurtIGaxR7qrEPfUeCj5tHXUAfJwYylDgmW3abWY3dxEuhWlpVYhR9WHyCU7iGW2rwiIWdx7APc+4+pmEYjlHTyC87UNBPYVQgcLiegStAJzVRCz03y8cByMH+zLxJzHsVf/buX/s2dvwYe+aP/mCnhzuiL2tN1bTjeOVyYSyVGdo9LS0Eb8qDqo7Qo9xcvKU7RYLHLgwAGpra2Vvr4+iYyMlAULFugKd38RYjQarYvh+eLl5SVtbW0SHBwsHR0d1kVR8yBHRkZkwYIFsp+T91SxC5/OmSNmV1edC1/rBVhour/S6Sqb0OG4BBVDWocss219LOktIOmUxGFxmRmD56lsUDpvtMl2AYD1hSAEMItrRj7g1s5MkAM4Lu7djclDg7NiQymgzsJ5W3SRhUtGl6Q7/1iDGvndQ0Nt613L2/IbwLNVdinPxhAOvwc3K0PYYQKd8T5xjIAxaU1bUBkx50WA0fMQH+ipuRt9IKgG5JoCAbORS7dli0Ke0TZPFhGptKAdo/2JDciUU2KFMMt2eg0tMExs0QwBIq30hoba1PuGDcouGRlG9JXqOKUPRXbDuKj3xULOYXqGO1m/B9dgTPSHw5hjyFmb7xl55GtwojlnVjEjcXUwjGY2RtJDFcjdVpEGEZEHH8Tk/Kc/YQK1g8zKzGD0y+x8LNCaqmcXEXmK8MfewsILsk9nbFf6Slkiwfw0iLJLoTt7W5sy2Wl1Bwha27ZSfrfan9pD0Y48liYf4AIbmEPMpaRfiqUs7MA9922nuUMd9NsKoSNe/Libmg9QEj4vaPSiB9tgVWjjlA3gkE0E5fMAnomF83LkssrPGAwGSeOH+DcJJ+5PmjRJent7xWAwyGeffSY+Pj7yxhtvWHMRh4aGRASFwG+++WZrEfHzxVF+aWzEofexEYfex0Ycer825aI9xbKyss8/SURmXOlmViQXylMMDg6Wxx57TB555BGb30+YMEESExMln6m/qtj1FB98UIE1CGpz+j0grrffVjyn6d2kCzJdd5sAg7BwGhCHOZcXozwYW72ad8VkkuXz7XhzdvZqPJhvO1w8LxHX2ZyjeDtswc0OBjRytA0QGjsvnL7Eljs7lhoEyVDkwIAdvb/8skL4IPeveo2id/YUGTKd4UWwG3Ps7VDHS9rgNXJ7WMuZ0YMWFEJ37IlzrpjzMsqvo7yN3R1oU7YbNS+JmfD8HIz0sldj3g9Y/ZA/PJK54fBejrbAMk8oV7wOrlJSUfE5yAj3Ia3vE9Gk8QjGIEPkrGrOEmCvV1c8n/Zl1CDOQ1vg2e+rQhtx3+e+x9wLJoIlFBIpTUN66GGHQ3FtJkaxrplbM9mCd+BC5bNnwVDPO6D0E85N5ML59jx0a3k9Nd/vUDhSKeY2gXiU64X5hL3ldb6UP0wvkLwdc29BFo0PtXGW74fntWsFoVDk/mfvx7syysOoAOch9lDIbEJPj3KftYBpd22ieYviPjtC0a9XzieSDEMHPK61MUaDpjIUnqIuhAKKyWXJRS+Kzs72OTkaZOnk5DRqC6crEc5TPF+mTp0q1dXV4ufnJ/ZeITk5WV555ZVRn9st85aSokwShOMcLAITUeuYK1dgYPBWNowC8EDxXkMMTaYw8ijkGVcb+Rw8JBzioAUQHU/azETlwc51HNdXANbVkB5+1pV9iEVWz8Jg1RUxoNESkIaB2Lp9NIyYbEKOWWZZmW29v/aasnUU60BdICOLAC1V76WBTnrcVYdnWB5OBoudvY3yBqA/bTHS7dBhxOSdW4oFhxc5ZunZmzxWJpFBpz0LWSC9Xhi99qAqXR+gi1caMMFFb8H7DKn7nbr86le4hL3yehrbmiA7basffieGNTldjoc6F01giHX2ETIUGedWb5CxEzpgo4oX2W2ZmFjXbQVsvTkFUPTSHEDRu7PUBY1W0P5C6oekUl7kd/lQuTMeFPQwp+Kha+98FQKkifrgMsDO9rbs6nnySTEbjVKfpIyvMC+CztmI57mCg3fUOHPLAZkeysSCfLAF0Os8/2OjrnewD/PAgw+in378Mdpj8naMf12lBI5dcOdQdVYfDMMwrAow/1naMOJXP8ecvbKL9E7jg3Wtzb0cHuEJlx2MXUTPuBy5aPbp6dOnbf776KOP5Mc//rGMGzdOQtkt+AKE8xT37dsnP/jBD6yLs9FoFF9fX1m3bt2o37m5uYmzs7N8+9vf/kKfxyEOcYhDHPKfLZddEHx4eFhyc3Nlw4YN4uzsLFlZWbJo0aILepSXIxMnTpRutphUmT59urz99tvWnTTOFzc3N+np6REjm5+q2INPDx3qEQ8Ps9z5X6i56kl7Mu7er3iN9qoL2So1JaJPvXGvsMOOZBxSs1Lp4kc7AP8wlMKeIsOL7GywkRnUAK+xPUKxFv1KKVeLLULyThsTAd2FbCdiAeGB/T6wMt07VG+LvLXexESbetc2d2ZyhaY/hsUK4kEIYE9t3Rx4kJuLwOhl2E0r5i4iOje6M1CxqCebqDg3WeIb9+Kd1htBMGhOgYU+4Sb0l08/wHCyVQFHR57i+mzsJZOrpdvrbg9Btux18Ytqngbhj71LllxwP8VGI3QWckD1zBnGYq+JIQgaCMNR8Ey4C3EX5zGxTmMS09iu94L3G2YhVIDao6AbZJHkI9QPqfNEzpo86hUYCeAKT7WrAEXy+PFrIsSBX4J1rQ0+cjcLyifT15+zO4nqce024D2WRsDba/aBTnkaY0LQ0lA8Z38UEBP2gL03rR71HiX+0MHMAUIpaM7pX4zn4teeHYd3ys4hVm9a+6hrsJeta7s4QKbHmqAzdpQ5TBPdoHqc1Ik2xmMu5fxfsbNJxMXKZRFtCgoKZN26dXLy5El5/PHH5Yc//KGMGzfuih7Envj7+0t3d7e4uLjI0NCQtdybt7e3VFZWytDQkEyfPl1OnjwpJ06csP4uMTHR5oIoouQP2S435pB/pzj0Pjbi0PvYiEPv16Zc0qL42muvyWOPPSb/+Mc/5NFHH5XHHntMJkyY8O96NhER60KneYMaI9XT01MaGxXroL6+XtavXy9PPfWUNb5oYsv7IuWHPxRxdhYpfBvWfhPFSjTSB1sw5kJg5r0mBH4ZatftIxaO52KrMrm40HrcGaV4cJPJ1O0wwlNMjYJVtG47rCKOlcyIgDV3sBjWXJMXYgk1amggo6sGPyS3NigJHkQSxUtj4gHaz6tCHLEvjjxF1VRtj0FcQKopP4ykp0fk7Fm9zjRvo9wH3uH6UniNG+lzji3MmoVn1m0B5QMyTF4R9BHlpfw/eYDSaWrIO4yCZ11mgndIzre49qC/BBgH6bib7q+6IUnkWR+Ahc6WPcfqagpx3BAHGruJkADSsHTuVKreTF5BRZHtWM79QeFiMJvFQMiD1cOkAOpwIOKqDNroQrbUP9ix4soqh5egr/T6K09t7oC1zzqQ4EAck3EbzopvoD8IaTlyROnjvFs9Qye8F2XLMpzCe2cWcpy6CO2+2Qtowboa1YsnL5WJYXalvV2kt1dq4xRPbCnvNr9mu/WwidANJmhVh2Oecfrer63Hb7yBc2KDibyi/nhfMeabhSZUipJAhL2OmhCbS6Car7MZnopaZT2cNQtjqdWikHQCugGRrC5GegQjBckV3dbjAwfwXByz1pVYVeGu5aHoL7uWkN7k0ud7e3LR8Ons2bPlyJEjkp6eLllZWeJrr7bWFyyurq42yTsJCQny1FNPyT333GPzd9/+9rfl4MGDNr+zyz5Vd2uoTALZhNdWDTmyU49bh4Dy71ItlFNmK0lORL8iaDgsw1Y0YqfvBXmhthQkhF4DSAgMn7LDnEAJvlbWIZ/AVFRi7mT7Y4BkrMLEX1KKnEWeDP3qFGLD+nIMijVr7Oh9+3Yxjx+v2zGhukoxfoYp7O28lQpQ0zNnD2CxZDiZhduMSRxLKxQYKbIGE2V1OaDU9m7bJd/cagBzLdwJmIsnrzB/KhKg6pJzDXXkHoYLmanCyX30crWBmLx4jzlt8khNg95ycmzrPSmpR1xdzXJwCUgoCVuU9uIFIiwQ+khfAX1wWGB6BxVi5wmUlJ3nBTgutU7hAvCuEAlFRMrhrV10qyUJcRjy8tEPU0VdTIgVzLt+uNWRcUbXPjoA+DbBAFhy3k4skJwvqPHk+HV5eBsMtvX+5ptKIXZtmDNBLjKY+gzD62SkLN8KI4UNYb8iGE3NiRhLQXWq8WUvcZNEt6kBkaQ614IVy4ZRyFoyybQVjeMg3I7UHvOWoD2YacxTHsPE2j13pBCszcxGrtx/hcn7F+0pFhcXi8FgkN/97nd2FxsRkVOn7KQOXKZ4eHhIT0+PuLq6iqenpwwPD0t3d7cMDw/LtGnTRERk8uTJ0tnZad1Bw8nJyVrxxpY4YI2xEYfex0Yceh8bcej92pSLXhRfeOGFf+dz2BXNkT137pxuwfXy8hKz2Szu7u7W0m+TJk2Srq4uuykanysWi4iLi0RnAWI8tR/wmdtWxaqdTtYPe2fs+LGlfSgG3kFNKT7nnKsAwvpStyr08p0R9GhpsNr2xtMzU9KVmbCH2XtW4RwyJ4/WgLqeoJlijHFRkHxmA7zDkjayCA8kWQ935uPd2ED0U6/DXpk7nAydnLr/O2Ixm6W6BUzipcsUXe9eQwF08mJ3NcEDzQgGTFbWh+fkahpsjW8sR8B/93wFjqmuADxdUg4yDFv+fiay4sno4rw3Pt6RORoT1XlU5Pmd6gbs6y1kWHJiK7mqX/oyyD3y3nvWQy13Lm8vlWCjgiksv/yliNkscpC2YzqaokLjfVBeezc84dwYqkRijMexvaLZBJ+kro3AfXyV9k0QIp6x20Pe8ox4Z1sf6zIDuH1TE5VOV1IBb2RmNxnydjZRTDAgVlIQAaUxgS51FrWNygqZtRNeGSM39iS0eLuYjUZpbVolIiI9zMcgDPRYIkhWsZnIDewYAKzKnJacNjxHKJGMgharfYj1SxAXV9biLLHOKHiHvHXdZAtuuj4cY2/jdgW9qM+BN8doCZfaO0ghGJ4kOgMBVW+Mwdxrdcc3UXIv59fyZHqYfncZctns0/9fYo9dumzZMvnBD34gd911l/T394/6fvfu3XYT++3Bp88/r2wynDofeYiNTRiQGpswuYHqitLExgw8dv2ZuaXtlSgiUlCBRNnkFTh2OqH8+De/wQLG8wxDtpEReNbqGjwrMwAzUpB3t/sIoJelFqVjuq0CrDXYYDspnFFVXvAZLnTrRhxj2EeBMHihtAefarsG8IbC2uQXMIBF8XATYmOzQ/GcTjd1W4/ffhs6Y4SI66pWdyFmGNmnDGDePJYzi9zzbUPfuo2KDdQfqHGyjZjUtLUtqIJYsJzflob4ojWvTMT+Ls4kh4oBeWmo8kwLJobeuDibetf2U+S+5dyh9k+GLAnC3R2ByWxp2uhxJ6LfZYT7Ief3WoNvhD3urgN8yZNzSBEmZ7uFGjh2YSMscKwKOuLQBkOfM70Aiec1YSynGm2X8dO6A7NcOc66daudcMHzzyvFKrROQdDoOdrx5+kNmJrXD8BgbF2G/saQPhtwnUcAx7d6KQZXAMX/uZMnZ8IgLFiMfrixBhD9+gh8PjgLn7PBoD0Lw/H9cTC4GEnleYNZ2tyM2Wswn1SrNXQ50sMwLhtLV1p3/YvNn/g3iEaYcXV1FW9vbyuxp6urS5qbm8Xd3V1uvvnmUb+744477F5z27Ztjur1YyAOvY+NOPQ+NuLQ+7UpV72nOGHCBJ21pUlCQoJMmzZNXnzxRZu/u/vuu+2WprPnKR4+rOQpMrrCEFxkh2J9HxbAq2xQJ88BZMW4xlKq8s8QK1tZsVmwqOq3K5YWQw/RiYAeKvfari4Tuxbey7EignHIXE8tBCtRQxzYwgzxh/W/fgssfvZ22fp3K7VtFbr3KVZeYzeC3r6+diznkyfFbDbLwUJY85q3NBgF74HhMoZl2bNj/bL1yAhxSTneSyPVsMcwO548IGrg9gFA5QxbcVkz3rnBbCLPSGM+8cPaiXvzTu5sRbPXzR6CmRivVvOZ3KFei8Wm3lev7pFx48yyeRUs8tRVSnvlLcPYyW1Cv0ofgKd4LAIIQ6wX+mQuQdFMTPE24TlDwpW2ZtR1RpwNfYlIvQ/uz+3ETjT3B41/war2tuAdK1vQJ7kJNNTg/B8frsL53Ke0vhbkj/fi/f4WL7bd33NylLzchYHK/Ti/kLkjOiIWzQVHqzAXsD5S92L8LfTFuNR22Fg4B3MC3+fYYoLEqcFOGfDe3gPE9KTO12/B+7q3KH2goAHtnyzknZL7vzsffXzpLKA4w/5AcdhrjO5W3ic6E+/IqbozmwiOXU75q5chl5Wn+P9TtBJy2rG2hg8PD8sPf/hD+eMf/2iT3PPAAw/YvebnBcCZys+wYWCgshgyfM3QU+8AOojFCwshD950f3TWvBY0cASVoQrTNtb1xcDk5+AezSxEJiryYtk/Bwthnonw9gNqpixj8zmYpbq6wOrkhZAH1AFil7bQolWjbh9UewSTUWb2Dpt6z8h0Ezc3N9nRh4RibQVghuOuCBq8gbT6FeHhlg4gDrp7Oxa3khU4Z30MPl8VqPwf2UKDt5QwGsJg/UxYIEubsPgZjTj2p59yCcAw9UVG3nrL+tnTP4E9uq4bsZTQOLBsM9bQQsENTJ203YCJRNM7w4/Z2bbLjak7dkltBya/vJ3qWCrGqs9bXO2jRPPEQFxr815MhFyKTfbSACGc8UCFspBEegEGP9WNMeNN8a8VxARl2I0XXB6H21KU8bN+OwyqNWvwjtG+mIRzj1ChzAgsTpFdWORnN1Eci6zkZlGvT7H8lTSWMtfbnmf+/ncRNzeRhYHK3+5NgDqnU3C0NonKn1EdvaIKjOdtazG+GqNoIfQlR2KO8kyLyZDiWO5gHGKRbsWASb1JqRk5aBsOc0yuoMR/tdMl11H8luYhNvYyAumcvfBCmuZTsYoien/VSOG0jjAD8Q3ya+SLkqsePp06dapMmTJFtm/fLl5eXtaKOZ6enjJ16lQ5deqUREdHy04OdIlIZ2enrcs5xCEOcYhDHGJXrnr41MfHRz755BMR0XuKCQkJEhUVJdnZ2RIXFyfHjh3TbTW1aNEi2ctmBYndPMX33xez2Swrs2DR7NhCUJoGTbFnxbgcwQoJcwDRMeFj9yYs1pxMy1BPdLHKBiNcszkOLM+gJmIwUg03cwys9d4Oem7GHW1VcSaG47Yu3Gd1BFuTYGzqvLc1sPQ58n1Gjen+4GF0r5077ei9qUnMnp6yKx/6WF6jWq+kvMp4eFPsrXM6F3vUk3fC0jwYDoYdexVWWIqxbGJJlVShL8zsI2+Sb8SV+xNBiOBukrtJhZ8ou9stHp7MYCjYp61F8BwYng+pw/3rQ0H4CBugvDv12TcK3n3VKtt614g20zl5nCEBTQiKbY3CfZkElVsOElT6HPTxQS+0KRO6NMLO0ih6dsaEqd5XQSCY19xM7nUgxugaVe0Q7YvRFuxh7tuC9x26/nocn0VfdcsnQhR55Uy806BUfmznGrxPb3DwhfXuq+qJqbPMHuG4Bu/4kYYxyu/F45JDQBqyw5715NKDo08472Ua4+FBMpSpseNFRD9xaQORHRRm6nNooQ2eeKsJ89YNN2Bj+JH3aC9dlcGUUYH+x1MZIyNzAZ5dllyyp9jT0yMbNmyQ2NhYueWWW+Sb3/ymlHLA6QuWL3/5y+Lm5iZOTk46KHV4eFh61K1KysvLrQuiFsyuYRrYeZKdne0IgI+BOPQ+NuLQ+9iIQ+/XplxSTPFf//qXzJw5U+6++27ZtGmTTJ06VY4fPy6LFy+WX/ziF/LNb37zC39ADSYdGRmR6dOnS5FqCXt5ecl111036vz6esUC+fDDDy/9Zm+9JeLhITuoOIJQZRjNij4YhZjPvBVklpBrcHQPRcDJm8vdD8s5XVBFRULhreU1KVZ+ahxiH0EmG5VoRORTdasgEZGdL5LTz94hmVGc1nBArTCygoy91fMRO2DWihtZs4mJeIcdRTifvTezumXXvjbE+DLtINpVH1wnJpNZ5xUN7lTihxyPrW5DjHPLFlx3ZjwRTcgTiz0C77BoFU5hcoKVQ0/eUGUTvEOmkTdvwgO2GvDeXbTFVgMRYzj2osUD65cgvWBwFTxfWQLPKKACVnx9OJVro3IfYZQ3Jsso1qjS+9fTvn+ZFJphSU1VyhpWUGrQEYPihaTGEAJAbR+QRDHOKHhQ6ayoLjxnCxGtQogNY3W0mS3B1aCpMyV341k2UjWX9T5wX3QpFGqMzy8H43TTJui6eQDvG/Sb31iPj9fg9tHkah0the+Q0EcxNM1t3brd+lG2UJsyyYRk924lprh7mRK3bZ8Dj8yvBjF/7zQQ+njvzXtd3rEev/feLdZjRiaYoFXgr44bI+KhToumWo8/+gj9l1EXI5HJuNIMu5zrLehcEepcmUzozimKt3vvodQaYiwFLMYE9Kc/wXNfSc+yY5PiwWb7EBmK3Nf6YKqGdIVy0fDpZ599JuHh4fL444/LEoYMReRvf/ubfPe735X6+npZs2aNZGVlXVbt0fPlww8/lBtvvFGcnZ1laGhIRkZGrPDpypUrZdq0afLjH//Y5m9vuukmaWJYgMQufLp1q5jHj9cnw3Bv0ALqBMuu3opG59sdSqIFj+EE6lCNFFQOCcTEnr5MIezwgrB0MU383OOpA+7YC1YaoxozCwFbNa4CS8uad0l7HvK162eho4XtxWAviMFko8vZpFVg9163Uc/dOzBw4U2GSYFa6baMObZhlgADILD6brRBWBcNGjsJaUeNmGy0scl5pbq8Q4blaNWuDcVixYsswzicyL++RSUR0eKxrwETOdd85Ou57yVWHVsdNEsOpoGgpEFKPLmtWGG7vxcUKGxrRuzmGdVJ385uCWy4METH1+AhY94L40VXg08bQwxDMzZKkOXqcPRPrmUQG2rbUNzto0zyzGo8WAFCzbxSYifyokzvXEbFFL5yDxAq11dfxflq32A2KwNU8+dfeFPtwUSlH7p14Dm35eM5VzdgsdzoD5LZ+hV476M1MOAS9qMf5MZh/kmPUCDd3mDAnmZaoAZXYJzrNm6us9N2jBdTjqW2Z6V7EYy64RSME4Y72YbKDcZ408r/nX9Ljd80PRD6HDZhvnPuIyPRjM8vRy7aU9y1a5eEhobKkiVLZPbs2TJwXj3Cd999V06dOiWdnZ3y9NNPfyHljY4fPy5DQ0M2k/d/8YtfSFdXl2RkZNisYKOVgLMljvJLYyMOvY+NOPQ+NuLQ+7UpF70oFhQUWDf9/cY3viFPP/20rFixQkwmkzz//POyatUq8fDwkB/96Edy3333fSGd4eabbxaj0SjBwcHy/vvvyx133GHNPSwtLRVvb28d+YaFNyi+WJn+i0fE2dmsI8YUwRCSVcGKJbSrG97LkiXkpVDFfykij2s/gu9sXSf34fN1WbDicteo19fhfDC/d5Tj3EKC6NhT1YV5yQ1hskaIaunmVSGtIjUU5pyu5ju5QMkmeGOtabDsAnxhtQbnK9bs0XKkqdiVW28V8fTUmYYZaxRPMbcC1ip7CVIFRfZRAJ/ds4A6QFGtB1AlJp48cI0YEbiCyAMG6KuxA1ZnSBeuMb0FMNp03RZl8daj9Wugy+YOxXLX5VFS8/ZWUO5pOOH33GHIi+f0lZVJ8HzdVM9rySpY3zvI4Gc5e1bExUXv3UqDl4iI7OtGWMCfKrYkWOC1xRIEVu0Lr5cd9FNpgLw5x89ZTblZuRaEtDiajQbScJ9EckxYf7sOoL9xZtrSNhWB2YsLzmPCxyx430w+SUy0nYc64yOQP3TQ61qlDY4Ewyu7QMllyC23iHh6WpH+kCOAKVavIheqCYP7qZvPWY/T0sg7rAJqw+M8hrrkxnylbzM6XVgI77CDSDQJ+dDkeh/oqY/ySRkoDCNPUQMv6uooZWQOCH8GI9qaQQ8OUeVtBQJU2wVdWzlH1Lm435Ztp8kvksbyZchFw6dTp06Vv/zlLxIeHi5xcXHy+OOPy3333SciIh999JHcdtttcuLECTEYDOLm5ibt7e02Y36XIoWFhfKtb33L5ncuLi5y7Ngxueuuu2TChAnS19dn9ShNJpMsWLBAchhkJ/k8+HSfCR1jYQwYdq1GJR4XIIA7dJgA5eRUCiYJnjOn14HVlpyfaj0u2InOMOyrdAbnUlpk6T6toVjEeALybgNrkXtuaz5YegHFgGE0dllIMc2aBGGd8kH8kRFbhsd43WYS3fJERW+rc3CNrCzbev/tb5Xyegz5anBy7n4squkNFK+hwXiwAVAXNYEO6ludD6bn5jlY3NYtUQKdvUZAYHyNIBMFQhnjpL7V2IVJKqQK7VsdivaN9FXbl2Zb7iPRTcR25JmaZzKCcueSEcV6d65Q342oiL1HjtjUe01Nj3h6mnWJ71o3i1wDtrHuvRnTIgOEIeeNaZQ/xp2fF/UmZSFf2UZtSmGG1gZMpgHd6NfVFrR1pAGf7yjF5ytrVIOBIObhPlzPuQrj4Wgf2sDOxhE6SDG1FJCmpmMuVsAhD3//C7PcrSsDd3x+CF456CHaQ9E2fjEBNs8/HIN4ugZtc93WHfvRZ1fG29Fjmm2YlsfV7D5isWpQOLV5rxeez7wV4aLWJXg+3VZpNThmw2RlqGokka4GfXFtvoa9GssXKxftKbq4uEifOnree+89mTJlivW7SZMmSXd3t5w+fdq6EHJ6xOXKzTffLB4eHjJx4kQ5d+6cWCwWa3pGTU2NlWHa09OjY6b29fXJ6dOn7V5327ZtsmHDhit+Podcmjj0Pjbi0PvYiEPv16ZctKf49a9/XebNmyePPPKILFiwQD744AN59tlnxd3dXZ555hl566235J///KfU1dXJ3XfffcFF6WJlz5498sgjj+i2hNIe18XFRf7yl79IQkKCzd9ezn6KdXWK5czF5NkwXtimWMPN8wEZBmUhZ6gkbZ/1mC1G834iS5C1XyaooMHQknbMFjwzGYnMqstNYpJJayCuzYSPXSnkfao3WrgdXgeXEmMvlC0xLoCcPAsW+OwUmGia0bh5E4yj3r4+25bzT34iZqNRz5jVjvnGhA8f80W+UqwBln+7Pyx/vyQcc6Pq9iJUA/e9ApiUYWh7e2cy14A/5/MzTGj3+ngFfWDO1a5424QE5/nEOOVENL4pVTjR4Ujqw2skDhGRgQHb/f2FFxQPfV4LILhTSxTPzXsvMQUJt67vg3UeZiLEhNyHgzVAB9gJ0hUeX6XCqky00e0KDaluQL/i/QZ10DZ56MvLFQ+d9cv3WbkTz7cjkN6THtbpvrusx6+8Msl6nJxIhA7V841cARSiOgd9sTc01KbeX35Z0bsuOqIKzz3crNZKQyL62njcJxjiIMSnN1Dx/jikMneAdEPQj4605kv35ImBoSLqn43FCkuYUSXe5tBtMZATnRdMmPhwODxVntu0Y54iGMZlD3M2uv5lyUUvirt375Zf/vKX8tZbb0lPT4+sXr1aioqKZHBwUOLi4uTnP/+5BAUFyapVq6S3t1dyc3M//6KfI7/97W8lNTXV5nfOzs5SWVkpUVycVBUnJyeZP3++5OXl2filSGam7bJXPU8+KWajUaqJYs+7UFhxCFqtGgcwSYSU451PJSHmo4M1ba1+ch6TakBZaBrbMBlwn8yLAdxpL2bDkA+PleQmTIAH1W1aeNFkMh4zGEsyseBmV2DBZRaZeQBQY+QsBY6szkLsLbOy0qbeP/ywR8xms5jzob/Dvor+ZicOjjpfRHTBm9y9oMxzd+DNmOc2YfI7lElJ3+rMNDsHCyVP3rxAFuynggh0/8gYQLzVexAnzmuAsaEZSUGC9IL6AaQXMPuVFxKeYNZ5kXFFs0BrJiBxDbniuhW9vbb7e1OTYgROFhu5MjTxFoRiPCR34V6nUgAlsqHE65zfXsCqjSm4jmYc7IijyZmg4tpMfM7pCJsDcf9cA+6fnkKLldpo6/Khf3uJ7SxsyPIm0scsMK509VTnK2GMzv2If05eBmMtMyzM9jxjMIjZyUmOlSp9Ozacnp30PrwK/ZfHGaPZ00sxFzTOwlwQspVgXm1l4hWKjjfHIfbOcwXfJ8gLC2RZHaDUGT4UC7eFP9MkwqUM+Z6ba7CKrTNgfmqdj/MDOpT2YLibnYOQNDKAK2l8X4ZcdPJ+enq6DA0NyYYNG2TChAny61//Wj7++GM5ffq0/PGPf5SgoCApLi6Wl156SZ566qkreihNIiIixMPDQ/z9/WXKlCkyaRIstrffflvc3NzExcVF/P39xcnJSZ5++mkRkcvfT9EhDnGIQxzyf1ouqcxbc3OzzJo1S6KiouTJJ5+UW25REkdPnjwpzz33nDz33HOSl5cnX//617+Qh/s8+PTll1+WhQsXyqeffjpqIfzlL38py9i8Ivm8Mm/VLbCEOP9KgztSfQFBNvoDvmVDybmc8uXIvNT2GeTriYj4FcILaJ2jQG0B+2Fl67xXgTcyfTGsYeYVMarCjLggH7y3tumn+1pYmCyVabBCo/NhtfVmwpozRwGK0t1UwzYIf+718rKt9+PHxWwySfJaXKsgX/XQKRm/0wTPiiFLvznENiOMSAevtaHNDnahzeZZFDShwAhEgl9jRpZtwomtffVERCJr4O32pgAtMNepBBgmntC7OT2AMTPy/cdwDuGtB/Nhw85LhOXe2je6v7LnZi9fToPxZhej/U9lKW3u3QEPYOTLX8ZznqXSW1QS7pAF1j7rL5ZYvZ3F6LeaR2Jv94f/+q9Pcf//ftJ6fDAGHv88YjbqXGpVx7vKAcUt7yY2rgljiVE8HcGqkGBVInUxIUtDERn1Ze8lKcm23t98UynzpiE7nJ7JqDhX3OMxzK+6uY6Kh5D3d8gXnqI2JLZlEhxKcqwB/YfRoYM7gSCU1OC9mQDDx+vjlTlv3k4gSQeXYdz1RmEsmdeCzNi8BnMft0eZEYRC685BNYT+EYy7oxCI3Urb09lFy0V7iiIiQUFBUlVVJTfccIPMnDlTJk6cKFOnTpXQ0FD54IMP5M033/zCFkQREQ8PDxGB58cL38jIiBw9etTmgiiiLKj2xFF+aWzEofexEYfex0Ycer825ZK3jjKbzfL000/L008/LadPnxaLxSI+Pj469ucXJRp8OnHiRBERufPOO+X3almzt99+W/r6+sRisUhDQ4P87W9/k7OqFevs7Cz7OSJ7sdLeLtLbK5FFxIbgIJWX+n8cvJ+QPrK+jiDhZ1sdrJy9hM3z7j9MZFlSBMvJVKr8f3BVvPWzSAvh5OSSciyK41+cz2U2wKI+dATW+NxZSkzjUCI8Qo4vetH1DqdFWI+ZAGTmfAB2EdQL9c+Ht2RP6gaCxWQw62JHgxbFXnOjl1qzH57ivsWwQGv3kufch9SW4mLyFKPgoc3roticiiYkFyPGIeU1OObACrmEyRxr5gAuEU44+G80KmSM6DmUU0kv/OyzntbjYzF4vliq1DGPyprtOgCrmz1VbS/I8HDEo9fbKfOmpYdKHALJXSp6sbcYE3jwH2B0zrWQd0bBwwNEBMvLoRgZETE4zqkNqzAD4JLycrTXRx95WI+rO8g7pDQMJw94ByO/hvtU4q/0ueWL8azHauAdruyjCk4HWqyHZmq7HYbVdD7iwAYTPKbaNAUxKQgGisKeYiNlprCENh5SKjhp16mAq5i9E59nxGPM1xpt54E2LkbMntGBhTHw8vz91WfmikyUAxhbhfEfmxmPi3gB4WJPlRGux5+ief/4cRHRk4VkAA/LfkpaFvo41z7jDCSx7LUehjWo452DujTAfCk940rlqt4lQ4NPbYmLi4s0NTXJXXfdZXObqLy8PHnooYds/tYefKpVr2eEi+d8DebQsegEk2N1jW3Hm2GQDAPBMpyPxiwZdUFvngMcwB7z0XstYJL2LJAQ/PYgD6hzGWZFZqDNC1UnGILxyrwAx8wYoMmDWQicqMQLAk2SzaIsYKxLk8m23pOSesTV1axjk82MUc47ZcFEyTpgW6W1jiZhZlTQCr+Nas4yo3h6vAodMcuT34mYR8yMc7aAADQoINpwya68cgxUTTXM6ONNi/2KAbu2JsKQYKhsdhfYzbv6wHrmogaTDer1yVrqTUqyDVsvWCBmNzc5tgT3jrUoENjMTbaZ0TzpM2zIsB+PD+5vTOLSEMlD5dBBZTF0E92Byb7EiD7JBoBuEYgC3HusW1nQuY+wwRhWjnHCfb93LSBWhgXtsbCDDmzWv4yION3wsfW4p+dm23pXCX25vspCzXZX/17brFC2cvsHMM+4zyf4lBaJ2eHoh4e3qrrhBYXjQjvtOAG8mXA4FmX3fPTD1nj0Qy0H2rrDjehDStxf2GfJaKHSCzQn9sbAsTB3KBZGpxdCLDqCGG/jcYX000v2FP9/igafasIxxZGREXFxcZHOzk654447pLu7W9577z3ruXU8m5wnjvJLYyMOvY+NOPQ+NuLQ+7UpV/WiGBERIe4qzDA4OChubm7S369AIk899ZR1Y+HGxkbr556ennLmzBn5iEozXaz09CiWIHtiGVGA6Tp9FciKrUW3ZXBvNnXD4marOKMIVrcukk2YxKAJFnNFhOIhRhEEyjTykBby4Mhr89sOGOcQ7d4+1wJIcV4KLrorR/F8lnfDqzAmkuUZSJYl3adT4HU1+IBwMmM/vNMgVYmNiVyAy7a8aFkgZidXER/gLkvXKAQN9lKy5wMmXbOGyDVkdg7vQRs4b4UOVrM7xTie6lEdbEOu2TyyyptN8A4HqA3CumGZbjmC3yYmwjtMnU8Qq9ppjtH+jLGriEZO9yw3wlNkB/1oC6zy5fPRBu1JsJibVdJNEGPR9spenT0rMjwssVGU9hKu9OcScgk3hmMMsOUfvQS6iWYcnwzSxE2w2s01IJ+lpSlj4ugqeId15PkFkgc0k3cEIfcwJI4exoKxFOurwJ1lFYDbZ/gSlklsjpJSePnLInBK8148a0kbxi87W1qKSRd5lT09aP/s7D+ITVm+XMRsljlqGIIdwlYjclQDDOQJkUebX473WsgpHETiO1yIco4Ze5NH3WduDCax/jVAknie4XQ093x4sPVR6IdhTegbWorOHEITZhswVx2ugefHEGvGKrjiywtxzq4sjKvBUoWoNrkJiIB3HCD+lhbqZ3JlclXDpxcq8+bs7CxJSUlSUFBg8/uAgAD54IMPbH5nDz598skeMRrNOlQzLAudtHOn0jEmb0G8QReE44mIJrmD3WjoeREYnGUdgAJmNADSqYxQOhcjApz7xfAUi5niafuOIGawUAB39KegQ7tvUQZD+zIsZgy7BuzBYGFEkZ8rmbbHGU7CAqnFHfwaMGh6o6Js6v3VVxXYOjoYE6S20TNPwitjEGOpN2FBYUiN4W5Oy7KyP0XEPRGDzRrz5JfiYCBhW4fnIA7CKNPkCkB9jBd3RqDdJ7cpC/rRbixQCXvRFrvj0EZsCMwrppgsBXM5MZ3QOytcqcWLRS6wO0lkpJhdXHQJ55ouGV2zgeyLiL5tePH2NiKW19mHGNnkLDKQVINWx6g1Qo8MmTJMyig3hyV0O3N0KYuiWygWDy5msD4K8WOG7AMsiB3qgvK08HTOR0hDg1U53uZWgcW0NyLCpt4TE5VwgWZ37F6L+5Y04Zln/r/23j+squtKH18gP654uRJEQwhFtIRQQw01jEFrUqLG+DCEUOoYv8aqZahj1VjqGIc6xBhrjaGGWh/HMdaqY/kYYxxrjFVDDSWOIcYhhlDHEEqVokEkxMIVEBE43z/OOfd9j9yrkUQvpHs9T54cL/ees8/aa++11rt+7BgaD08CyyoZL85IKAnHXsiTeXA4z5Ely3QBZcrzHuYJYmUsmoRgxXp9zfJcMG8ejqZmD8xfWm+nw7CuWdbMobz0S8QwtU6oLt8qMnpiKSO+B3RT2ae3m+69916x2+2SlZUlNptNfH0x3Llz5173JIwrnDp+DeXn56usMC+Q4rt3SPHdO6T43jepz3qKPj4+8oc//EEmTpwofn5+0sGYpojcd999HuOKnjxFs+0Ve2WEWrgMJK4rYqciow6exLI6WMUr4lBb0z4F9XAMVTD6tHS27vGdqIO3xwkrbGTx+EY1k8XHxC9EVp7PA/rn2lvFrs921CGrkS10rh/iJAQey2g/wJsui5PrFJOS3PL95z/XPfRFDcgQdLnrxKRdNnhWPLUrYmAVW8xhwsHztwG25MQGM0GES1p5Xhx2QEg152CURUUSNMpYOXlzeWGAb02LedFE6m7ELi654vWJ8Go4G29eNDwc5ziCi4rJUzUtd5okT3xvqqgQR3CwZaIP2HRvPyWOvBQPZ6PWdgCu43FyS0IOCzCqst2moyGc9BQxBR585TZ49rHFlBjDkAVlbpyOAz+GH9Xl4WIq5CW0AZ5E0Tl4EuMrKBOZBrPuEL7DeWW8DswtJ+IkIML2ZKADntrrvfiiLu8LC3SvqGg1PPXx1d27OomIpCQABbLA/yzvtAdy9ruJmPC8WJrIU7IgbyihifBaL56D979rP7z/qXXIXHXtLZyeztfshbLnyZsIL2zGWM2F6qaloYhIaxwQmNvWENwbNGHCBJk6daoUFRVJA1fWisiYMWNcLd6uVYgict0TOjwFwMeM0VPUuZA+lrCAYqODPJdVbFhL8ZjdIa7LRMLVpQMajZVfxmEozpGcNWkkKXPm3iI/CJ/rJG2xQhUlVYh9MLzH+0gjwXf+/sYFBUqmhwOzn+4HAa0JA4zMCpqNgpjJuHeoba9+QRoob+pUt3yfPNnYd3fT5mts7s44wCn7SZnxmpKyaNdlezQ2M4ZueDML2A/IN81QnCE7wTuLDiDorDoOWXVR1WAqx2SCbNhglnADBwPzrm1DHC7iKJRcZQw29bJi/GxeOjbDA9QOq4zefymVm7jmkiA2T/J+tuMuCe5wSHgqYO8UM2azc6/rs9rZMFYipNbt9blzMOAY3ZNcCGhJPPiXYPA4go5h25IFRWgvwy1iOeWVNsKuyeBHdTG+Em6ECEJXY17WhSEswCUF0VlYg4wKcoiC11jAYSrdMXFr6prB29SGDe75Pvezn4sjMNAVDmgjXdFBDR/Wr8XnKY2IiS9NBp9WJUFZbd8NbcDjME+cefWNQNdnvsvREMGCvxMmvXYt7U9JVEJCvYM51nRitr5H8Z7AoajQzWTc0Jxub4b8zQzHwHftRbw3IUF/ZjPN0ShSuEEdFHf+glHFXg2fBgcHy4wZM+TZZ5/tBp9evHhRBg4cKL6+vuLj4yPLli2Thx56yPWdnpynqEiRIkWK/r6pV8OnIiKPPfaYVFZWyoULF6Srq0va29tF0zQZOnSo7Nq1Sx588EEJCQmRS5cuuc5TDA4OlunTp9/0eYpnzuiNqdnSYThjh5EVyB48W8WMhjEEx8Xi807C+mJP8NQaskANOM7MuBIRCaCsK0uxGGE7Ow4DzpoeQ8X+lJ0wsoGK3hcYlhvBq4WN8Mwm2fH80+GAthgRGRmODLmZi/F8E8VbNhH3cMbHX/dcv+EdFCw3vX/G5Xhili/HNUGF+RXwHhYdAoRkwUfJjK5N1b2XiKNIXNnUCK94TgNq1zjxID8c0CjTonOUhEXu+ooqHTZnfjAjW8dhrFzkzCc6VO4Hf2LLqJ6NIK9Jh/UM5MKdSFpy+vldt90Ys9h1LiIhF6fr4CWwXLNnxSL5ve9dcl1rLf1c1wy7mWvIUl5WQTW87NrTvDO8/9BTSLp47gfYxrYsOGF9iIjUVuD9ORG5JIegZ87MonXVKhg3Z6abECQjJ5wPEx3toZ3k22+Lw26X0yE6smI5s5MYMn4NZLmoAF554Ul45bxGT4VgjY7YSR0bjLXSRT6Q72738sNw7HEb0JPRNsD+rTFAO4JmwMszs3euDB3q+uhHNC+8BEcfguduccWJgaOygDydWKujLturMSZGjvkVJtGy7wn1avhURD+X0el0yuXLl0VEXJ1zrl69KpVGy4hGzpISXfiud3SVqh/yDim+e4cU371Diu99k3q9UvT19ZX+/fu7/m06tv3795eoqCjXd7q6ulz/5+/dDC1fLhIQYE2PL4oGxj99so5xm0e+iFhT3yurgYFzXGodJd1sEFwvoRT/h1fDKlxtWIhjj8Krs+S/kzVZHgbLmS0nWbAc15TgkFtN39moBzNqJlPz4L3486RkeDLDT5JFbYnhYlyJifAUXUNMbJQb0Ztv6tb2lCnwikwv5GE2wymxoiYE1moUIQIhlKTCgceUbNybE21SCgyPj/hrOVooMdp1ub0DSVLHKDZs4TsXspEHuyzHkJPViNksbYM1v4o6yJTFIBa5KBXeYb4fkl+2NMObzayDd+oqMeBAsgeKiBBxOESWVeC9XPEx8sSHU8xpfxX4vjAG6MbVf/xH17VGTTTkMOLSFRWIRT3+pG7chr+PdVqbgPcYOAhe4ID/+z/X9fRmksNnnnFdhvNOZgaTKYGIvW/2FJccxZjqyGPeLnBrgmgxt9ZhLWVmO7rdz1O5FFOVPUHsdofEhuvxwMpzWDcsvwyGHKeY7aQ2TqzCA0esxjrmhJV9ibpXxvl2UZxFxwuC3nV0AXnu5Hg0zoUMnFuN+HysUdLS2UJeO9eb5lLJzWIsmqWHsYdxyc2Jo9RSsE6HIs6RWMfGIH4fW0UtleSLdbTp9fDpoEGDpLGx0aXsTLrvvvvkt7/9rYzyUJicmpoqb7zxhtu/3ajN28gQqqfhYihzU+aNmjJntjRAADihittzZexFRhwLAAusmaQ18xA2qx2pyGBltCHqHOCTSVTsyguVh8t79uhoA7ahtDTneozVMYVwCFrtRalYLJyZx+9gKhZOOFuzxj3f9+xpkgEDHJZ93ORNUDYW+oF0BOpTJpM88EMYwyZ8b48dfGelt7Rav39aHe697xCMG95YN5VC1ua0UdYdZfRwy6q8LCi0Ux2x1w7JE1pnTRrgOlhKe66PgxJlSM+3TofZ6v04c/n6WZBMC8N0OTudBNkbfhi82dCB+ZgXApnsmkaKlci3wH1LMBPBZiUSUEGZuWRQXEzCJhd6CM+0WJ4k5EV+utx6SMy0dALcMgOG53E7NufRVXhO/US82xBuJGAKKU8eGWge60P37BHHgAFYyLT+tiTCUM5MhhG0bj8yQTkvhs9qtZwKMpFCESa2yNl3HFpi7JvWz+jFkLHjkYBJt6dDETIvTUOYE9wYHufP09aC13zw+bxECvvQnG4pHXm9oVpKKkdTT4yeUK9OtBERGTVqlIQQzmzCp3V1deLn5yf9+vWTgQMHyl133SV/+MMfXN+7nq5X9UPeIcV375Diu3dI8b1vUq/3FK9NtBHRC/O/9rWvyfr16+Wpp56Sy5cvi7+/v7SRxbZ792753ve+5/aenjxFV73cZCS1nLZBkIcbcMe+wwi8pxUAxloag+D1qjZKuKAo8J4ZsLLYSmZjc6TNsPI81BS0HgK8FrQeCR/OuWjzxh1c6mPgQbLXaCIsXGLCpUHsgXgwhi21TxF2sqJN041cI08dbd5+W/fQ2VM0ER2uieTxsOXvqKL6SHLR2VuyJHQkUmKD8dAj1K6OrVHuqrHMDg/5+DjML+f/cH4I1+CZfEo7B0/gRBK8SkbHmdec1MXlaYxERJ2kJC0TcyPo2BkX595jee01/bQGwukOLNctdU/N1z04ZxJwkuaATHiuD100Dl6AeYI6e8utWSg1cq6EJ+7YDL7vigTfGfVgQGdRki77znjIPROXZAy+FzDtHd//Pv7AmRtUV1A7Dc8354l5FbsS3rBz/frrJpaZiAWDAUf20xoiBnNoJtZONYv8MtSmZt8UeOjmUmRAhTvacNKW9uSTruuzf4VqiGojz5PrcmkC29OnXvuRFK6lBEFKLONSK0cjIXPcEYraSZpyF1TlwTPuIOTmCx6o2OuVoif49IEHHpD4+Hj5r//6L7e/+853viPFnN5JlJub6zYAvnFjk/Tv77BAj0ENNGFGml7NFCwM3kB99xPWTxmOfHzSgAEYk/YsYIOMk8jG2pNtKD1SKBvacI95NmTEts/A55wVymvaAlEVo9DYXM2jlwOeYgjJoi1ZI3GBLWmEU+moZzOFOKoYizO3stIt399/X1eKDGum5RjGCK9eqjiuzIZy4ezJpTkkJwR9lgg2yLHnKPPO0C4HmgEVsSLcEYd5OT0DMUDmLytulgeGuc1Y3/Y1pJAJAVm1Bpve0ix8hw935bZfzjDAaeYJAiIiB6piuz07L8+9vLvCBR2k0IydvjXBvUIJqsMYMldiDFvSSfZpY8ujfpesUE1ZZfiN+c4yOzqSlAATKwS2KkwtRbJz7mc/c11HcHuw5ZhT52LMNccgOWvSciC38RKVOViPnH26bp17vh84oIcLTMOGZSYoEUZ47WEoFE8GqmM9sqNLkrH+xu6FgewK+9APU1Zifg+sJYVH1uMuP8DGzI/CeOx/Pr8E/7SXdD7si8Hf0yK7y5aIWOeLBYM1N18bm/L0YsD3bJTxVhVK/SJ6Qn0OPjWppaVFnn76aRk8eLD4+PhIv35I/fbx8fFYjqFIkSJFihR5ol7vKSYnJ0ttba2cPXtW/P39paurS1paWmTo0KHy4YcfyuDBgyU/P18KCgrkvffec/3umWeekbw897VknuDTn/5Uh0+XTYPltKsMWYtTow34h+AhbnXFHsPoo5S5xUqdTPiMHNx7Tw6gpeHTdGjh9G5YWQ9nI8mDPaMhG2HdHp8Mq43Pt+PAPFulAeuNMbLnx5lohAtVhgDuMDtkiIi1o7K7ruEEtTiXL3cP4z3/vDhsNgtcvKJZtzYZJpxjo3ZuBGl12ZEswt4yozwclJ+XBL62x+t8ZR5xAhQ/P6KUvCG6oZkEICKSaaNEEPJU9yTr8A43ULfwjqzlFVlAJzhZhOc0PwfepCMGMmiyhX83Y4Z7eT97Vq/LdWzEOqmfrXsYlibnjIcxkfvQPhsWPM8BQ4OW02eid1kHfO1zyMurDYH3ZIHoST5P1cE9MGFVXzqguagUMuIpL4uTNfJKgZgU5gDRmZQM2Xe26d69Yzk8o8xGrPu1az3UKZ45Iw6HwyXvfGaopUaautVYvCyGKYjZ+xqBdqRVE5xorm/G82mt7epA6IA7brF3yPB0ShtkeEMdfjsvVZfb1rAocUdBU4BI1W8D5D/Ejvfcsdd9TeikRl1eiqiz1vgwSsxioRvrHuX4vNTrlWJkZKS0tLR0q0WMioqSrVu3yoQJE+RrX/uaNDY2yqVLKBoODQ2VTz/91NIFxyRP8GnTr3+tx1g89Ho0d5oDpdiEPGUTshDNiaSYDylFJ7Uj4piXuZkXrcHmzb39eH2E2rFIyysAwbniktcOkgMxRqDrSDaEnGUrpZQg3XgoXIa2WEGWlAZ0+45jNSCdXF9fD3VbF0TEIdrblHlmaKP8/TAcuHUV959lBcAQnKfi3pQpWHg1FfqCjFoDWGxSBTYUS0yENSRPGN+cU3BZCEzDiGWLW12dwybMsUOGhXjeGVXm7HpzU+WTKnJXrXIv71VV4ggOlk17Ic9zRM80PTUOSm5EtXv55Uko34gYNmdvx9JRWpWLqc2XMTmxU2BQVO4lXhM0uqkKvOH4HSu0lGKCCw0qmgxlz7FZ31zI5KRSwI+FE8mIJnhvj2S4+1hGNuphji3UXjEzERt17q5dbvl+8qQeU4wKa+3+Urw+ycjkPqS8nEPT8WyLTLJGMzammkbsN5wLwLqSFVFGFfixKQT85X2O1+TI/TovRxSAv6cWIMzBC/JIM/azhxNJ+ZNnMWkB1r65L/H6Hh6Dvb3wEMImX7R4v9fDpzExMTJw4MBunwcHB0t5ebn4+vrK2bNnXZ+bSvDixYtSVFTU7XeKFClSpEiRJ+r1nmJgYKB0dHR0S7RJSEiQp556Sp6hAt5raeLEiZYyDZM8waevvKKfkpFWBQikfgagkSELdNd9zzQkamSUwipiK69rBrLQfMMA7bTXof0WO79DSskaN1yC0CxYqIxwsjXHnh3nFWU0wCpP2w+rnz2M4XW6dc/ZqUPSCXogaKt8MixFhm/ZyOVi44BGHd7jRJGkJA9w0t/+psNJFAfeYpvX7Z04s5O9BHbIGMHdnoUs3doYWNRsRJsokuUMzcVU/Mv4H5noF5MxN6FxeMfKo4A1Y08SVGpM9r4wJEalJcCjWrYZHhV7wUycBcpe8Ilceo7hytSGwQOz228A4xHVd+iyynI1PB332pEDT2h6Anl2TJwBwQLCAm961PSgHYewTqavpWIzmuB2qqvkuWaZ3JCrJ+YcKEP2MXs0PAxOqrOEP4RQC8Y0+ceG188HacdmQ3acO3e65fvzz+thmiUzjAQi7m7Pi5sxfX5ZGk/JZKA5YxsAeXPNdGa4vrccD8PYPLWk5LXB6D5/P6qOeMOCYgjonMWYo01TKLGP4hm1O7E2IxogU5ZQRAIl6ZhIC0En+ybDC01LpGSsCMx7T6jXd7QZPHiwtBlFsHa7XZqbm+Xq1aseoVEmszXctaTaL3mHFN+9Q4rv3iHF975JvV4pxsTESF1dnVy+fFlaWlpcHmNISIj88z//s/zbv/2bDBkyROrr67sdIXX33Xff1LOamkTa28VikrOHNt4I9oQRdM+A/IkQxD5iyIByEBDOR88MIQu4djG8U7M59cUKeB37jsEb4WQZTrJlT62kGt5hImWuD8/CGOfF6fDyhvXkhXNSELkjHCcaGY5o/K4wWMmWpuVGweMkylqBbWili42+0tHlK6EUHMsMr9b/nwtzlZOexmfDopxDZuwcyqI4Rc2DRxQgPhIfD6/XdQamjdwzMpF3nMRzpk+heA/Nb8lezNO37wUvtTcRP5y+W/csd2yE55CZDe+QOtjJ2CwklpTvBE95rtk7XHQUXmt+pG7Fs4N7jTMI+ugjkQEDpKgDfBq/W5/PnDZY4VvIjZ6+HHNgcdFp7bWGI/7FXsiocMhQS6B+jNF7bwGomn4McV3LGYNUh9oWg1jU+MNAadpSERtcul73FFZlY144D4BFPMIG5GZ/GTzV0XYK3PEPaL37fF0P6zBYlceJaiXU/J1o3DjDyTS8zvpsjN2y31RgxZjyIyJSUIDrtmK6MZUsldnhKbYbbSNHC+L/m7Yh/j9nNj4/dw6fc6icEZuEBHjxDApMMs713JSF2PuOKgT4pgs8xYhw2nMaoYYywyjBq9pNrJVkMSkBfz7VAO9wxBdzFPsufLpu3Tp5+umnJTExUd5///1uv/Px8ZGf//zn8tOf/rTb3zzBp03vvScOu92CqZRsxqY09piuuLaEAFJl+MWRTjV+tGEsWw6PdkUSlGJ5JOAMhm5SzhnQJ8MnvMuRsmpNxj0s7cHcVemLWFMYTe3KyR/WNv+45qp+hnJ4jHwfQxmmpePdCwquD5/WN+C75jAC5gJutGh9SiponYxNImgnasbyGvBb3quC4rFpL5uh192tyKWMWlLOI+ZCYTB8y1Dm+GjU7nE2zL4qKDczG7BrATZ+31Js/HnF2Gg4l4XRNJ5STpTYlQv4aeFmXYlbp8I9300Yz90B0pzsw8+yHK5MiUedn37quu731lv4DssKZanUjNNr4KJC3DR8uPZ3tMjMjM9ryWHD/NU2BHR7B86R8pRHN/Uo1SCSMWc5lSKkexuysdQLlGH4uXPd8911KsxmQ6mzcHImCQnZrgbsLVM7kOF8cTJqCUP3U3Y2Z8MYhuK+SBiwabsR3rGkafNGxFYY7z88RjISyv10g2VkBdUBU4ZTeR0MEz6th/ezmXORBMePNG0uTiqbmooEnQPF+F0KRT96Qr0+0Wbw4MESEKALufl/EZHYWN1iPXr0qAtGDQwMlPDwcBkwYIBomib33Xef23uq9kveIcV375Diu3dI8b1vUq/3FJOSkqTU8Ao6OzvF399frl69Ki+//LLMmTNHjh8/Lg8++KDExcVJQECAlJfrVvPAgQNl2rRpbov4PXqKZWXiCA62pIN3JSHxxFcMK5k7aVC9T0kVLKGxNgoSUw0Dt2tiaIlvOadMt+i4awt3hC85BluG78EOIZdNMCSxrBSwyoo4w+JkM5rz/j1Q2kaYYlvfQJssn88gSqGbDbiS2oE4Rdzy/W9/0+vlfA91TzaauhNeIHssow8DcipKAozG7z2kjuqY2ANmb9l4TvvchW7/nNYBmLJ+HMbCYsVGOTvUXPPle8yA0qhkw9IukHjKNVx33on5ePRRuIqF2cQrzjIxkoG4FWBHh3t5f+01PbEsJYRgPsMUH70TaAg7BrE7kdjBLuaeUkDB7EVHVRNoznNgZr7Q+uGuQiknAXfPqQLczU4Vl6v89rcfua61K1/XLwjz4/Mq+X2Y5kwDj7h8weKdJAECPt2hvzMvmRF2/N0ZEuJ+n9m4URz9+0NwWJi4JIPdTg8eXAnBxt/+Nuq0X331QYy52PAQCV613I9c6i0CdIWTk5ako8RrWQEgdEYyTCCAHX6ecv4uX7PnPiSRahw5icdA3rhe+8h6Wt+MWJXT5z2gXq8UzQbg19Kzzz4rK1askIKCAvk+9ywkmjp1qrz66qvdPvdYp7hkiTgCAy31eQfGQejMDf5ENWIPLM+bEqgmh3bndWVY7AtnIIbBEOCpSCxaM0OUj7DiEwai7HQPThFjHIwWV3s0hDhgNuAWUxluiUc8M3MtoCDLrsOFXhQ/6rJhY/cNwUayZa2+GfCiOHHCA99NY4TuO3OlPmY2Flgp8nrJiCRIi1cY121xBiHvrKYlwZsE7XJLG6AcVk0kyIfT9HiXJU3oTIYBYiaujuqgsRIdaQN86uGUMEkLg/JKWw1jjZHGi9m67O6KQ11pebkHvptZv/QQs0kDF7hzKzZHA6DiAxWAoVOiEWY40QaPaFQMlML0uZCPHZsN6Is3aiI+dYM35yEdyDLMzAWsuSUbG+HxNl2GR9spxs2BeILhS6ZA9sceJoXPmywT38cIupmt9UREUpoBHeaWl7vn+9q1ulI0oEdLyz4/QIJ8uDMvP1bSEW0E3bPskzyfEF2RjDrpvvkFGya1jXhmxBrIvoUftChPRXfHKkfEdIeyRazGEi8332p6B9ownH7YZx12wyng9HnuDc0bzUyChntAvR4+jYyMFD8/PwkJCZHk5GSXkkw0OGyeqWi326Vfv34STXGwzs7O2z5eRYoUKVLUd6nPeoqzZs2Sbdu2yQcffODxTMUJEybIYc6QM8gjfPrUU+IICJDTuUjWYGMkqkyHIQ/4wQPgDC2uHeNas/oCqtXx0LFiQyMs43mzDWvRQ+Pt4XSYKCNnnIgxQtx3p2ePxHSIJu2kZBZyCziBgL00Nu63L/BQs2S4O12RUfTnG9SHTqTOFgb0dSIcluioHGpVwQPirGOCJyub4UnE7odHYLaQE6EkKLKEt0+ERc2eKosS84A9mahSwK1/olNa/vCSvswWTYRHU1gH/k4KI7idyVNBJn/OeK8JV1vOyHR/WsO//EuTBAQ4ZN1yIA9LVuvWOYMOzAM2yEeFkIXPOD4zh7yycsoQNJdBQDOebTlRo4DWNKEujJ7MSSDZ4+cbi+8EHd47qpjaLhLvahIBiTOAYKnnrSaEgFycwnDdI7HMHXndzmnT3O8zb72lJ/QZ7rgl8zwEa6CmAV4bO4GDBrW4rrXvAiVr3wnZC9iGOmWziTaDSuzwMvHUcacbC8JBe1hlHbz/2HDjXXkuWGYpg6ywDegZl2nyfhbQBt7VNuvPYT441sKz52b9w+F494h6fUmGJzI9RPNMRbvdLkFBQfLyyy9LWpqutEI9tEv3VD/kXL1BxOGQ4dQLkjec05v1xdFceu0vdWK00XJAx04ov+nbGCaFIjxEAttoQBhTpkCBZe/F9en9pPBofHycEdcMHBf89uEECJqrEDoZGHB9I+AOzqx1lS6ISB0deXSgAffmDbOqWP//zAQogTwPba9efVXE318kzU64kKFUR4Ugrb52G4yLiJ20yXFrNVLMsVWIu1lg0GkUcxhnLFraJVIp3hEqpDAEkGnRMcS5eA+YdxSr+pvPP++6/vpc83dQhA89BoNv5vdhm25fibjUhp2Q4aws9H0MCMGGWX4MG+nIBl02dqVCeZR7OCUjN1cv16hvwzPyFujPrhEYM+PLwOvyiSRjHDQlA27XRGzIZbTh8TShHSeezbGopZOhaFbNBT/mNLDxQNsXGY2VSbqyGtUGuJkzxjNtkAu2p/btBuw3IgHr4NTJZNf1jp0A1/yM8R4PgQIfnYQxeaxT7OwU6eiQ7Qk6X1PJmOUGEVFsbZEi//3vadcPgfHDtlGdwHDeEa3H3IsEoaDYvbTH0dydFMgN72EFBVjnoWuggAoEimlFovEDgmZrqpELEbUWczBpMkIUk+JIcTZk45p4UWfXlSjH7A/MhnLmMMMXVYq9Hj4dPXq0y1v09/d3XT/00EMiInLmzBnp37+/OJ1OOX/+vEshiog89dRTt3/AihQpUqSoz1KfhU9zcnLkhRdekNmzZ3s8U/Ghhx6SI0e6l4x7gk/Nc/3uvbfR9TetkyqfTS+EXAP2lDhBpz0H1lTAOGpZRZjEjnhYbgy3hlbpUMVCss4YLeSGwAzZBtlglW3ZBnsnMwme5bKdSIIwf2tppswZFmQqVnbA/OLz5TjRiNHTUeF6QsS+UkBmyckeYOtFi8QRGCh7EsEP02vggDyjyQxvMUIzMxJQV00Marv4Pu7uuaICXhgXSPl8HZ6M9gwlXdGE7WvGczgZpj0RyTCm0R+xFx53URw87vF1qD3jCd6XADli6Gj8NiQTdG2DdW+ilSwX48a55/unn+pZv+zpmg4zW+QbwiHXm8Ixnjl1lJjCLhdDBiwUnOVsPpQGuqkRc8DrIaqDYFpyCUauxPfLhRLETIiDXU8ax6lseLIj4skvYEaw20WD6YpBUo0pg+zQMewcHu5B3j/9VE9wMjo2XNwM2DP0XDnfwO342yOxFhkS3Z6MsA+nRDttOozsSlYRkQAb3rt9v/vwzvZioAUzp8GLZjSJM2+jdureZ8007CfMm1FVSEKauhtzxwk4vLc4BLw73aDvw8OFZIHiRacaqXj/C1a+9HpP0RO1t+uT5OlMRRGR71FMhykvL0/VD3mBFN+9Q4rv3iHF975JvT6m+MQTT8i+fXqCi6ZpMnDgQGlqapJ7771XREQcDod89tlnMnDgQGlpabFknDawmfI5KGaIUxwOkUceCcGH1HWkK1H33NriyTvcTWnOFITjM/kip+Eei+ywUj0dwXKoSr//ukZKLT6GLx9uhpXFiT6ZCWWu62nTKFHBDzn+K+oQa5C9xv/ZBeHuFvx5Iy7ZM/NdAG9nxzh4QaNCdBMyrQzehPtiBBH55jdFgoIsaecmcVxjWQx7U4gB2mwI2nNsqTkcHhwnTE2Nh+c8Ncl4x82wkI83wBKnIzplzmbEYTaRZ5fWsM51va4a9Y4Ld+I6wkQIyLQf34DygiN0ynkcJTik5WK+loVDdsaTN+vbASt+RaLu4dQmIIzgoduY/OxnIoGB1kZBJp/YE1+2Et7hinTEeKWRetNxUIdbkXC9CF8b3nDtOMhyfDX+zE5bhWA+zjXjmkOa0rAN14bQVE7DuGPToZBGLCBvjMZ9ohleYMxsXP95INCqr10AsLZosi5HXXG4N5dLeCQ/PxE/P9k3W/cQ1xMbc3Ph8T5cBQ+Sj6/KMDvhiMj21cgO8rn7Adf1W28hyWj8Id1zO5IK+W0vwxq4GI48B566vFzy1M4BMbOcyVpBvDS8fhaFjKOII+5KQmx6VzYJJQWb2/0QK5ejuNFw496n6zD/pSROU2M41uw+8fLzUq9Xiq+//rrl301NTSIi8sknn4imafLYY49JV1eXdHV1ydWrV13f8/HxEX9/f7f3XLJkicwlP92ENQqPOSQoyGE5h00akPlmLtQRYdgYlpyE4vrFLBQQf/YZbhFaDOE+YMMml0KHww789N9d19pfv6tfFFMEnjC/VCoZYqV4vIMC/msJ6iNtczwLG+tosxclQUWLyvA+SfT4qXYqFq8gqIyyNtMT8PGRUl14bXTw8ZI4p1u+y5UrIr6+bvsGcLKu2Kn/GdUJTp2GzJiLlAhSV4avs8JtpTq6oMXdGyUcIgR52UnatLdR+6os7MgHZkBBTqYaQ6EkEzNjOZI2+5FtmNP4eMA/DH1W7gdunkOb0aq1+P7Szdgoijbr8NL4Y5A5T/Le0iJy9aoVAhs0SN9cnn8essTwFltv20vBx5nchJPxfcb33Jxay8bK2CRqIUdYuTMaisKxmIy6RCiEi9EYb2h2goiIVJDStB2CEuBs2tEFMFyil+MenNnIinDIXqyfZef0saxYgP3gHGW8euK7LF4sEhAgaYblkRYNoeCM75KVIRgbLmVhMdb2Ourp++GHgC05OVlK9fXBUxETg7k7kIMQU3Y2DMxFy6EI2YBn+Vy/HuOdKvr6yEinlokh+GE0t9ebS6m+xKMAWqi1cTBqc4yvbE/FGhzOL7mcjHnDieop9Xql6ImGDRsmf/7zn+XMmTMiItLS0mL5+x133NHtM5Py8/PlecoMVHR7SPHdO6T47h1SfO+b1OsTbdLT0+X3v/+99OvXT65cuSKBgYFSUlIi3/jGN6SqqkpGjhzp9nd+fn6SnZ0tv/jFL7r9zVOizTvv6Ik23CqMreQT03SrbJ0NFpmlQw2bvdywOgkWDyeLjG5GUkh5GL4zss0AGmkgx08Guf2dpV8SY7AcpOdsGDqOYU+dngjCngIb80X73XfXGG4nCI1+fKoZgfkRxbrn1TJ/vuuzzqYm94kHzz8vDptNChPAV1fjZQ99otrtSIAJOEnQCeNuRCXR091+xZxfnnOGh7n28+GjsNA7/x2e/av/j056mAIr+cRJJCSYjj6fbck1PEtWurfKeep2+MGL55ILnvZ5dYZnTi/p3LzZPd/feUevlyOYvName50RzWjrZSES4JTNgPQYseAkGW6c8vAxQmBMd4PkdFMxIEumOWHwestjMtx+Z+RJeOsrqvS5XjaX5JRamTnnQs4c1QT/ES7I3VS4o8+WZAqXGChJ+UmkZoy0IxHEGRbmlu/r1zdJ//4OyYzW1/GeRqx9zlfidclIR1Q4ZOzAYcgYd0LiyEfEtlWW8YqIFePkgkT6/HQ0xsUyxqU1fBsTcud6T0Z6ODKzbjX2ljnZ2FsYtmfPdsdqA9XixUnJR77Z8PivOTvipqnXK0VP2adbt26Vp556SgYMGGCBTZlyc3PlZz/7mdvP3dUPnTypd6+PCsOEbSrAhJmT5LsZEIpzGjY5zmpkBCmtFBDiKhtgmaXJhKvzAcUh+oLkzYXlluELFrohQpsAKxPW7BzXMXCkRRuxGXFMwXIUFGXyLTuErEqGjuYtB3S0Ya2+cPfsx6L12ObNbDfGGX+GlhqeDEV7Ogd8n1M659qviojIwnPY8JYINmHm36QyfH48Wf8+L3SG1yynQvCk8grnzYZ3UHofs+6P4zG+bZCzwqOQs0mlBH1zOiM9s3454N4hVZCjzM363FgOzB19fb5zL11Td3CcMaICRtiyYmyULHvTU7H5HzgKBc9GBZ/+YSrR2EZEmvfVje72dxHr5sj340xFXhNmATofQnuCamtHVVBsmg/sTYdc8DxFtcFA4JZu5v6cvxpKqrAY8n7kiHu+nz2rZ/062oy1Q5plQzWaVXB8ko8WS5mI5+Wvx/NYoXAS+djNenOOmuXITo2y0V5BP5xaBdnbNYXCBcT4FWWIV/PcRBTrfN3UDAOUtxveknh++T3rkxGuOLUczx+xUVd6R6Ygfv9wB+Ry0X7IZT6VMPeEen32aUpKivTr10+Cg4NFRGTUqFFy/vx5efLJJ8Xf319+/etfd/uNn+GxjeNVqEiRIkWKFN2A+qynaJ6SERoaKn/729/cfmfr1q0ymy1tgzy2eTMa9Z4aBy9kRBUFbU3zmTAhbrzM1ipDHEVHYc2Nb8b9SsJgcY2thvV6Okm3tDiO7JuIJICuUsCFvg1k8bF7wD/m1lTUWswcL2fKMnSXMg484lQ/c3winvvzmihbYTwSX5zLl7vn+/nzuqdIxZjTG3Urn72Bhw+D15bCTbqe2QZvcvs28vIIpjuwDTwzrVTXiSEiVs+a6+yoB9aeBHj8GeNwv8IyeMsMeQXU6fAPt73z0L9dls5AB5f2cHw/IIe6yTBWxs3Ojck8TRmbYWHu5X3fviYZMMBhcYDntOmW+BY74Cj2mibZkJTBrbrYu43aiOxI7mjCbkOJn/5bDiewtz6ac5U5FMCDoZ5gGTvhYZghgNBqrJMDdVg/KUdpfHNp3FUUluAaR8Ixl1VA9k1i/nF9Z2Skh33GbPNmvPzojZmu7zDczB4Ur1FGRhjViN1M9ca8kA0vr50OB+YQwshojPHhVHj5DNwEncR88NmfS+KRgLekWPdy86aB75V28J3Hyh4/e5PsQTIvTASLz8iMyEE4oX0zYG06YbBH1GcTbQICAqSlpUX+9re/SXh4uPj6+kptLVLcR44c6VYhilyn/ZKiW0qK794hxXfvkOJ736RerxQPHjwoq1atko8++kgaGhpk8ODBUlhYKKGhoS4leOHCBVm6dKm89NJL0mZYdQE9MBdesf1A+tscMoPLADqicW2a/mTmsMVlQWsp/tOW5L7zh+XMRbqnGauYk4xYRvk2fHfkYXSgqI1HjVFEHUW9yW1buhPeIVvmk6p0b2zDasTEOB6UvxnXizbjfUqXw1peF+8+vorXoSDlCuID0dnPgiS4PUiiyLXaYZ6qTSZy+xrEE/i0pkx6V7bU5y1AdCB6Mbw5ezW+43I8yLPOyEHcaM9sePb7EnkecdkVBu9w0lrEhKQsGdeGW17nB89vezxiWMuaYeXvOobvcB7VeHJJ5rUhcLJhNsWmDVN7uI3Mcg904YJIUJDInEZKgDG87swseC+F0xCL4jMP//EfUXek/bDadV1TjGST0UnwWI8nA20YO1lnYHQqYkERNkpao7IfPqsxg/pnHp8GHnC8yvRIQqnT9LkkvENrLrzD2eRQzZ6NscxcjDkdGQ7ZKV9D3V9MF2ctZZawq+OB3rmSKAP8HGI3PDdugh21H3HQTamIg3K5SmUV5JrX88kkzONkAjuCjFrquRvh4ZlHu4lY1zwjM0HLIZPtK3HvcNrzyiMh73krDXRsLZJhjoXDU5w5EU4LL+C2JDxnUSOhQRvxoPHb1oqIyKb9KEWqi4F3uGwu5FW2UGefHlCvh08zMzNl69at3T6fMmWK/OpXv5K7777b7e8eeOAB1+HE15In+PStt/Ts09EhlHlH91hUqiuD/FxavARjZSTj8z02gllos98VAsXBC9nSad9MHePqZMZMGDoj7GFpxUy3X+eMtpk7IcS7ZuvQx9QwwEYnQrAxcAZbrB0C3RUOwfTdjWB4ayogrAEDzouIiPbSK67PnFlZ7uGkWbPEERAg28dBwc4M0ZXR9kZAzDz2iwWAbSwtycJoUTGcTHBSUQgyGE1dOLwACs/nOdzwv/8bG5Dl3EbGfxg/85Cmt6lM35AYYub2ewequ7ffExEZNOiS61r78Az+QJtvbSJ4ZCrRgMWAP50rV7rl+/nzesIHw5aVawwjgGR21XpsmozKM7I8J4SSMggLXlcMg2zhAoJSzXVFslxSgYzPsYdoHhlHIy1gOWP0EGTHNR+UBsmwte80aunHuCRRud8ot185sJ7ajBkLi1u/cSMFZ1ubW74vXNgkgYEOV3H8nsPgb0YH+FgSiXGOrUMG7q4OyO+TTwLa1QZTW7bJUOTb1+r70vQF4C8nyIxfANkrXAuZnFQMmNmSWMY4KKeaGrJfmAuInU/gmd4GZcURikXxbgwNEYtsXEzV5zq0jRQr47t8eoED/OwJ9XpP0exccy0NGDBAwsLCxMfHRwYPHiz19fWWv3uKRYooWMNbpPjuHVJ89w4pvvdN6vVK8f7775eHHnrILXzq4+Mjd911l9QZlovD4ZD29nZpa2uTDnaPPie9+abe9mr0RuCgRTuhbPOXG1afH3kJFA1PJsOlPBmJG2xdNxDsN3Uiw0Ww+Kav0a3UHTGN+DsF+9ftByS1kNpb8XOm74T3sGsGJQtR0oJZAzd1J9zKGHo19hTlMCz0thB4ikFkuR8Lg2X7ox/dJSIiB+IoOaSEXDqm1atFHA6ZeZjGabjRMQQfVq6FdxjrBwu8sRGWYWE6PL5JE+GZbNoMjy8rHff0Ndv4Uf1LU3b3EgURkXXnAD8xrDm9jmqnCH8KjUSZxcVDx7v/MDzadZkSTrJAXujHHwfjHcjrmjMDVnQVASIR4cY7czx9D7wMpkOHdPi08hB5PwVlIiJyKgbyw56S73pA2GydL9uIuV9Baf0J46iOmKFF02Wm2oGNxwguXIx7r9qPe+TkwIOLopIWiwtuukEExTDokrgW3hg7PaPi4eWNXA9odvVqyHBXNNaeb5WOKHGrQ0stiwf62TN6O8miUl1uM9Ihp/lrwcdFCYRY0fqbOhtuVvIFeIdOKrPY3kweVYPu0peWwlNkx48PNAwLoc9JlusbsCaGcBISZdqtKtDHMo2ArNqV8A53NMILXXcY3mmRH0JA4xcTTEIonZlLNmHCYNdnb74JL3RSGzlFX9BT7NPw6ebNmyU6OloaaZJ8fHxE0zSZP3++rOfTK4k8wafPPtskNpu1eD+tAlj6nhgd+2blw/EAhpPWraTMTU5DpNW5ahuUC9fRmc/neOWoMghXYSSEgaE2RzEpFd6AeJBUDb5qry6Ylk2P4pWbqiGscyZi47wYgo2B0Q7mi/n5vGQsBGdkpFu+l5Xp9aGcybpuigHBMLbHOxszm+E12ggrqxFX5kLokWHYMHYd1eeAsz9HivuC7szF2FQ4JlNDSpmT/lihugs1ZawHVH1gMSBs5imPm+tWPSUamzaKtZbVQxbkiy+Kw2azBmjNwDgJxZb9iK9xRMKsRRUR2XcIvOZ3/fGPG13X2n9Tdqch8L4h4B0nFOclU1tBfhkaa2sWIOKgKpozw/DrvPNO10dvvYltjo09rv+riUT9rQW6W4vYFSN20/caCoyLe6lZr3PKFLd8r6rS5X3IZEPBk4Yq9EOIYBIZAPwdjuONjIdC7aIKO9+jgDDbk/Q4cMBygkMZ5ifj5lQk1vyISIzdPORXRCSiA9nRmw5BKZtNFtpTAe+yb8KNKBbVIY54ei72WLYZWY5GNhiyw5sEXW84hD1pHtkoPaFeX6d4Pfi0qqpKNE2Tb33rWy641NTx3/rWtzzeMz8/X3Wv9wIpvnuHFN+9Q4rvfZN6vad46NAhj9mnr7zyiuR4CJRHRETIJ5984vZvHusU9+0Tx4ABVreBvKzjSbplyjkWI8uo7RN5NZU2QD7cof/AGnhOKbNhgZfshfs/9pgB3ZBr0BqNe/C5ievWw67h4DV7DxFlZHXT4PNKdU9lSSTV6NEPSwSWM9+bHXDzpAARsUBYW47qyQfs3G3bdv32euyh1YTo/IvKoYQlsmgP2GCNsnU5uhhW58UsWKMcoJ+zHB76JiNzM/8Y3pVj9uy9cIlc/m5YyNx6ijuepGyjhA4TiiJ57TqGxB3mU9pEwIJdNkCwnBm4KQeeO0N65tjZKrfZ3PPd1VnlEODEU/H6mEfYcH8+vy+gGEiCMwlehYNg0j3LMY/sWW1KJzk0PDFu7cZoyfAOgg7ZE+MMEXYlSK670nXZYKSFvUNGWiMWQI5q1wNmjqCG6uz+M/pgLpWAjQQp0/icHR3u95mtW8URFCSZh3ReszfFoEdGBWDo4XQyxuk1GNv03Rj/jvWA4Dfs7J5UE1BNPKVFU9OM77Ijlp8EudglkGWun+SMe7Netz0E+xojGrxOeRp3tQGqt+C6/AOTSKDKE4GYMfojERHyRajXK8UXX3zRreKbNWuWrFy5Ur797W/L2bNnRQReoojI1KlT5dVXX3V7T09t3pree08cdrslE5DhIjPLkRfbiA5sABuOYmOYl45J4oN209ooS4+zSKkwfHuMHhebmQyYgqWLldXYk8i6W1KFzNZ/+wUSjQZduYL7UMyjcvaqa29ted8l0/B8s02ZiEjUOSoBIKy5sg4Qi6lAguYiQzA3Kso9343epxYsy4B3uGkBKyhW0pzFy3OzMItiTryS2egx8UkyBgoPw9DgTZ1/Zik0349MSS7bSIuGbLTH6bLBIadVIVDg3PeVE/oqt4HXBxox7ykx2OBiU6FYKlca8kW4a25l5fXb63FgzbxmzcFMYGvLkxXGk0AyPmcBFIq5KbIxwDzlWzBszEhv/mGKV7LiNKFBDxDhiQbI8qgYKC2nQH4dBSiNWFINPI4NptF7dUVVkgrlxdC30+l+nzEPdzYjAJbsaSHsj+L1U8MBh+5aQ/sCWagMQw5fjb2gcIq+RzCveS2xUh5+GPtJ1Erco2YljP/aiVjTEaUUsjEfQEI+byX2vg05GHdhBeaADRYWNW4nd2q1/pwRzZQBzgLD6/thlA31hHp9os38+fPlscceExEdEh06dKjs3btXQkND5e6775b29nb55je/KS0tLfKXv/xF+vXrJ52dnbJnzx755JNPPJZsKFKkSJEiRddSr/cUDx48KCkpKd0+nzJlijz33HPyzW9+U0T0fqcdHR2uRJs77rhD5s2bJystaVY6eYJPXQHww91bromIDK8yoCOykM3m3SLWcrVYOuT3VAFqEO+7D9cXLuA7dGCC5C03PByChObMhfeyiYu1Cd51LkfGnMPmvknxpFxkUFYW6FYXOwXD6+jeBF9krnYPczGCtTSerEYzXYySjJwJCe7hpDNndI+F5mpLvP4umUKFuMyPYszLpnB4aqdnw1NjC5xhIfbEdp3Tva8T6/Heo+zwwmpseO+oEIx92Rr3tXvs4bDja+aCjbcRf9mtIAi2ZjMgSj5kmaE1Pu+vfi7ef0izAXsS8uCxPvTTT3W+k/CVJOveD/OOvSOGvQI64Imv2wyYl+sRLza6LzQ3xWNhGzUOYM+T3RrG6EgmR8yFR8BwsekgslynHSIPjF6oJgbJTtyacdUarBmWFz6kISNB53U5HXzMDuvGjR7CNMbpJFtKdU83MwRw6OkEwKGcZMXbGH/Oz2PEhPlhzqVvM8ZSXg35HXkOsHZtAvbaiMVYY8ezsSeOTgCfTlWBT66WmDyP1CC/Zg2hZERR64GSnJoNeeBza/cc1SFZRrUWpsPzrKWmGF8QPe39iTb9+/d3+/mAAQPk008/df3bLMH4xje+If369ZP29nY5ypsOUV5engqAe4EU371Diu/eIcX3vkm9Hj5NTEyUDz74QER0+HT8+PHy0ksvSWhoqFS7yXM/dUpP/GhpaZEPP/zwpp41+DerxWGzycUFsLz5xGtZbwDeZKptPAfce944xJC4KW4zHeOi3QP3oaQKHklePCXsNBtBFMrpDwuDpZu2GrGlfc0eav8oTnTyJKyoSWRCurrUUCBxzn68zyY76rO2jKM6lQaUQyxrQ1o8x49qEvT78ElKI0ZQ5wqiqouhYm93SCy5RZnpRgxlImWXUPBhkx/dmMbfQOExjt9xBQcnzMhqPUi1n2JVBY3wDi1lZ2SmrrCTJ5MMz+N4PLwGzhMwvc8jdZi7h5MomEOBnahiyMJMcsvXUQnPDJLRIQWU6GHWJ7K1Xuie76vzA8RmC5DoaCRxzIzTkzVyc4GAjK+Gt94uSG5wdsA75OSrhbPB7Bkz4JEcyMY4Hq4y+Mc1LMRfx2agHs5piCNdjAHScWov1s+ek5gzruBwPdsPMcLwEHxeRp6f3Q6vh71y7reeEYbYnjNM91TbKHTJ55GuQqjRQl0j4qXL4ZAKU+YSMffDTwJtyavAWuR3YrliZCItDvxYNhGDmjpNl7ldc7FOGv3gIXPQlro5yjLK5Rgdx516yDuk+tpToo/Xj9bava/90HWtzaWSHHbjKcA5Itl9UqK5nTEax/tBCCU/flHq0/Dp008/Ld/5znc8/jYqKkr++te/dvvcY/ap0b1+zmYsPM6FMfcbFkqu65u0Btl4DDNN3YzPeZc9HY4NcvhcfMfcPBim5E2Hx5RmJ0FjvI5Wy55ibHAZxaTEzMXAAWsuzKVNass5jI/hHm6ZxuWQG5KMjZ2gKk/ZeBUVOmwdEQI4bulKfcPlJIA86l/qtCHDjQ+KnbcRyRe8QbFyzsumTDWDsRcXYweztJLiH5IRtmwyNmpWnHxGIsOShUm6EmNWD2nAom+PwWYQ0EiFyAQjlsRBIVnaoBGPW2P09w9qw2bl9PNzy/e1a/XDbufYKfvYtB44u4UQF25fmFENxcW7VWE6FBA3UFixEsCU+RiG/JbGELxGUPmSo1AOeQluxiriNlPxRDRkc9RGJI1sSUIyCUPDDLEva4OhYCakiVgR78yJOnzXGgajk9fp3Lke9pmf/lRPLDNg4YDJUFA8nkdfQbLcD37wA9c196J1HcYtYuUBJ54YQmfJYD9KhhJhwpz8xQkwfGs2Kh8ug0F2OlXfWxjiZKXN64SX1ZIpyHSu8XNvVF4ZoPPiju9/Hx+S5TJpI+bagw34uanXe4rXg08feughCQ4OlkuXLnX7u6+vr8RzFT6Rar/kHVJ89w4pvnuHFN/7JvV6pXgtfJqfny+PPPKIq81bYGCgR6X4xBNP3NSzqkISxW53WOoQ2YqRBhM2dY9fZGXBm2Ivb2kZTBc/TjlOB8TKCSkpk3XrmlsrLeqghIRGtGpzUlPkBdwE+BAs6gzCf6bWwbLLDtH/P3Y3PiuMgyeZlATL0kZoIUNeySRB49uoDi3OSI5gd80DVVaKDBggEtEMi3XcON07SBHcs7CMun0cArTLpmtDAzzF9jZ4KbV14GVeAWDI7OW6FxC6lvhLmTNHFsB7eTgGHmQjJTi4WsWJ9fT4c9ScPDFE/z9b33Nywd9N1BawfiK8sYoYeId2hohIwEzvUARWeizjgh4oNVUkOFikpArPG9to8Ju9MIK1M+IBy+2LwRykxWGeJu0F1F8TB6+REV3zrM7CY4BXh89FLdzpMnhYeedwj/qJuPeQY5TYxWcuGlAKdYqTvXvhHWbOhlchScALl+VSzU3VbNdlrA0JHdXUWHx7sX7N3v+SuRh3LokU09mnciQ42CFRkbp8th8itIdgoKIsAvGojR6faSnZVK7Gmw7J8JZjupxlNtDv2GGgDKilR8Gb9p1AhAKywffaXMzHujLsF+Ma9f8/bMN6yNmG9cDJP2mTAcc+PBHeISNsM+vAwGNv6bwYL8Qr2lsK4zl/hBCMHlCvh0+Li4vlkUce6fb5rFmzZNu2beLr6+v67+rVq+Lr6ytdXV3Sr18/uXz5svj7+3f7rUf4dM0acfTvb4EhT1QDehy1WhfMEznYKEeFE9RGO97oiVjsLKsMG3AZ1YgybIpHIvVNirMaoyYiZmIp5qGKbs5a5HqupanUAmsGjonakq1/npmIv3Pt16IwxLaWnITyzYt0X6x8ug4xJtfxVxSjdBYUuOX7iy/q7fUY/R0SYiwa6o3JWNs+wX05OZFjDgzjMD/4gGAzc5NL9Tjj010GqYhI7FFAWLWTobgiUpFRXLkTRk9sdWG3wZZXgV+cAbi9oXu4QERkpmA++IQISzF6m27ExU7GRlNa6kHelywRR2CgFbMzhZIgtUVtgA85tuWp7o1bjzHeeDEe2aKhZfrmVmIDdDj2KDbBLWHISOThMapbcpigaiaDIUeOIfbFOoAzSNlIYRnIbIaMnxiHjZ9DF44yXcmYbdRErm1ukHRdeTczb2um4V0ZbmTlt6cBz8iIR+ywPgT7Ass7y7A5Zwx7MjTJ7QjZcBkdDSN7/DSEK7iJBN9nzAQd4vzxD6FSNi2n/ZENLb7mVGcKOZhtKEVg87Dtk1IAY26mH/bP7ZSe0RPq9Z5icnKyXE9va5omnZ2dEhwcLM3Nza4s1ICAALcKUURvv/T888/fkvEq8kyK794hxXfvkOJ736Re7yneiDwdETVgwABpZmuEyKOnWFYmjuBgWVEAK5vrB80g+kw7YIXyGMANI0uppo7bn7BLyMVabPYQBrOoQPc28hNg8rROgWcQNAWexLrJ8DAsHVzIfWjtcG8xnw7ToY0lye6TRviQ0cosWPE8bM4StJjxhll66hx3uXHP9/fe09u8jThHEXITTmYXhDIhOGmDMwVHtuFd8ooB3SxJpXZ03OKqTbeAOYGCPfsh5+Dt7aqCF8hnULLVO7UAHiwjx46ced1uXpsE2eEkI56kkgLA92PFfQ2pMxzegjmUiDDOFnR/rp/ZbmzSZkBzhed067y+GPziWrj8MHiN7YuRjBJwEnxasR98skCmjZQkY7jd9VMAy3ny2iyN4mdTyIEm7UgCvDlzuWX4EbxKblItHXwbkQAPaMtqeEbcw394OM0NxVY2bNRlcN4M8JbJKeKW77/9bZMEBTlc62gsnYyxPRUo1MxpmMOFi7GGec0fsWMvYNhy6hrI/q7FxueM8zJTmRjKYneTEnfK0xEWGLkTMuDKRLScfEzXVPDpGwYErmsB5m5LAjz0zDbAtKcn63LCHXc2dCB5al4D0B9ZRkloPaA+rxQDAwOls7NTOjs7LZ/bbDa5fPmy2994avNmtl9iOMp3ORhcMnmFcW/8nbPNtqzH4vlowADX9Tf+8hfX9abDULhz2giGpLhfZY6uXFmeOAtz6jHEckqmQCnxBhTQRguVhLs8Gpu2icIyZMMKJq0Dyr8yHhs4wzBcSBxKMSFX3JHwx9wVK9y3GzNOa1gnWBwLEw0F4AHnOZ2FzZmzXpfE0UbIOBnhqgtjsKmYw2PdW7KANm/GlgjHO56ABTk6DIrLMr/NZDAYgrIoHd/Nz6Z2XVydzTgtHQGVdxIbIO9drECi2gxojTRM7oYN7vluNk1go82cULJ8Fq6E4li3HFmt2/djY2O5YViVIT2GZ9fDzQAAE6dJREFUuU1ZDZgNCMyymHhDpvGVxIPvPL1My/wM2SDZKU/C7xj+47VctJbCDLTgihqhRHmtuHqKhlHZyU6su5ISD+0kN27UwzTGGuEGIL4NUMz1Ar5zMmlaGBlHvAB5rbBFZkzO0sOAqlc1whhZYofyYX6syMF+dqoaUP+IbTCWLZNgTLx58sy190sbB9nZdRjvbDlCj1+ULXhjfTjXQCma8LWIWOXlCyrFXg+f3ohCQ3Xm2u12aW5uFpvN5rZ+UZEiRYoUKboR9XlPcf/+/fL444/LnXfeKRcuXHB9PnjwYKmvr3f7G0/w6caNet3WzDiq/WEY1LAe2cphaOnENliaK/YiYYWtJSbOtAqtJljIsPiOVCPT7eGNZFFTGtfpBlip7D2w58NGFD8zIlG36FISEAw/MMNDHRhdL6mDp8rGHMM9Ls+MrH/n6tVu+f7aazqcxOhOhHEg6YZiBNvZQ007SZXRnIlB47wYDQs/NAG8ZChgepLuuVH+kaTYCBplN4y8+QPJee4+lqUTITv10YCwXO3XyCPckAi4fd4xwOPs8tekw3tmOeKEorHVmLMtbbqcZDZS8fsN2rz5BIIf2juGpU7MzsgG7/bkuC+kD21A8kdGjvuWgFyTaMoq1+5v2UYNtsgDORUOD2dEMbyarrnwdngdmrR0LnkglJx2KgFryVODaQ45BOVSpjPBjpUVekIRe8Msw35+7veZJ59skoAAh8tj5XDE8AaMx+dB8PGTT0Jc1/w8bkF3PBUQIrc+NNEQPleBE6ZmpmOM7TbsJ5w0dDERmfWhq+Eptq/EOgg4rCMw0wuAaPABA9zcntfp9jrce2YHhaAIZbpqIG/+n33m+qzVhn2YQYYl5Mj2hPq8p2gm1rBCFBH59NNPxc/PT65cuSL9+vWz/E3VD3mHFN+9Q4rv3iHF975JfV4pFhUVya9//Wvp16+fnD17Vp577jkREZk0aZK89NJL3RTi9ej/+4cqcdjtsuckLOCMbbBGzYBC0jZ4ElwdIZsRTElPh6c4MoYC9WSunahGTDGU4kgpubqHw+UASMgWS154w354w/ui4VWcsOPey3KRIj91GqzxOqPu7kgBWdTNCPzsKYWH8L1/QU0Dd88bGYnfbtgMy23yZN3DG851JWQ1M124INK//7XVF/rvmQehuxFPYD7u2Yt34rhqFNVqWmIf5EbvSDQ8qpxtrs+OrIfH/3ASYoAn4uBhJHBquJ1iG0dhGg8h4VjYoFvu6yhza95ilHKcyEZS1ahieHlRpYjrspfSuBnPrInEuDKrdNl0JpJ3k+ehFeAvfiESGCjas/RZiM7wXccw9xzizD+KtbEoBnNfWA2vhpPThldQ7SoF5KaX6c/ZUYz404ZEyOn8CWdc11oLrR8KTPrmIsljKWeIGXzfR51w0ig2POIwwQIUkztxEt4h9VOXFeyGEjRiinZsGZJjfAY95Lr+938Xt7Tp+fPiCG5G0kAiCTnBOtrrmONTjXgXDrvx0JyJ8A7nUJcnWaMH3XfvhifJABijIev3w1O02eDBzTtGiSz044AKeo4BBXA9IstOVyQ6eHHIc0469rCMuVgTe3bD43zt/+mAZiKFv2PXYm8+VgcE4YtSn4dPp02bJkeOHJHPPvtM/P39paWlRWw2m8TExMif/vQnt7/xBJ+a9UM//vF519/++7/vcl1nTNR/02WH4DDCyDFvXqMbFp92+4cjWdgI3cWXOZnTt4zgVVY0vCoYq+JzDpux8TCEZdbuMSw5NY6EnF+Idwl+PkGRlkM/O/TxMoTpCU56/309+zQ2i1S/qcQYvmRm0xg2bEMSwLxqwk5odZYcwnMZ7jThLkuSUgFBOJyEwowiw2TdsdFuv8J62ExEmWQnCIm/zIPiHYswU2cIFBUbEHPKsDm0rtE3B86mTUq6Pt95ak05ZDgqYj2Uz74kwNZpFVSdTtlO2xcAApzZSMlkLNCm5uQ5JeyR6xQz/ajwjF68ZiMUbpSNQiXmPRm7Y4ySsGeLzO6GUrmYDSUQakdYoL4RinNInb5WhkyEAVy/HJuzc8YMt3zft69JBgzAHjK+gn+DuXSUEozP2WSUfMX821ABmHnefqp1NYzA1iT8nfenvEMYf+sxrP+gamQg19gRxuAEQD7X0+w/mxFNexXznfaTAxVISEtZg3FZsrQIGy7J0tfk2Egkp53uwHrgre80bbc9oT7vKe40FsmCBQvk9ddfl5aWFvH19e2WjcrkqX6ozZWxiQ45ra3IIjUFvItqk3lN81m+7RRec3LHHfpDSwsWDOuf1lbzefjMlx/Ef+CHtrTgmp7ZTM9hx9l8ZisZ4k5PWp6fw1m9pBSb6WgaZ0ezMVR89vLLL8hqN8Ef83dOTrE1f8e84/ej+16+jN85eZxk7zGvefjmLZmlAfwFdxOjD5q+4nT7FZYB13N83M+Rp3vL1auuS6ev+3dw0oNajRdhVr3wwvX5zq9oPo6HxjxtbaU55h/Sert82cN3+L3MMdP78UAs9/Cjl6XvX7pE37lKAzafwzxlhhDzLDJL78ly69cF/l66BKVoM57T1UX3oHt74jvz8Nrf8HMtY2ZhYj7Sdyw8Y74a32mle/MycdLc8Xc66DmXNPcyzvuF+V6WPYQFiebXIkfu1r2I272S99JLHfgu78lOp0hwcLDHcr0bUZ/3FHNycqSyslLeeecd+c1vfiOPP/64+Pj4yJtvvimPPvqo299cuXJFrpBUfPLJJ+pYl9tEZ8+e1csARPH9dpLiu3dI8d071NTU5OL7TZPWxykuLk7z8fHR/P39tUGDBmkiov3rv/6r1tra+rnv0dnZqZ09e1ZrbGzUzp49q4mIdvbsWa2pqUmLjY3VmpqaXP/dzL+v97fb9Rz+97XPvFXPMa/5eebfurq6FN+/Iny/me/erudcy4OYmJhb+hzF99vL9+s90xPfb5b6PHxaYQTvr169Kp8Z6bovvfSSxMfHy2zG3q9Dvr6+EmnEd0yX2+FwiMPhkH79+lksjpv5942+e7uec+2/zWfeyufwtbv3E1F87+t8v9n3v13P+TJ490V+q/june964vvNku+Nv9K7SdM0uXTpknzwwQeu0zTy8/MlISFBampqbvDrG9P8+fN7/O8bffd2PefLeu6tevcvOq4v8mzF99szrtv5HP73D3/4w8/93dslc+5I8f3WjKlH1GMfsxfRH//4R01Euv03a9asm75XU1OTJiJaU1PTlz9QLzzHm8+8mecpvnvneV9Vvvf2eVZ8773P7PPwqciNT9K4GQoMDJTnnntOAgMDv5T7efs53nzmzTxP8d07z/uq8r23z7Pie+99Zp/PPlWkSJEiRYq+LOrzMUVFihQpUqToyyKlFBUpUqRIkSKDlFJUpEiRIkWKDFJKUZEiRYoUKTJIKUWDjhw5Io8//rhERESIj4+P7OUuybfwvpqmybJly+Suu+6S/v37y8SJE+XPf/5zj5/3wgsvyD/8wz9IcHCwDBkyRNLT0+Xjjz+2fKetrU3mz58vgwYNErvdLt/73ve6Hb11M/Sf//mfMnLkSFeB8JgxY+TgwYOf63m3gu+3m+ciiu+f556K74rv3uL7zZBSiga1tLTI/fffL//xH/9xW++bl5cn69atk40bN8p7770nAwYMkMcee0zauJHyTdDbb78t8+fPl2PHjskf/vAHuXr1qkyaNElaqHHwT37yE3njjTfktddek7fffltqa2slIyOjR88TEYmMjJTVq1fL+++/L6WlpTJ+/Hh54okn5P/+7/9u+LxbwffbzXMRxffPc0/Fd8V3b/H9puhLqXb8ipGIaL/73e9u+X27urq08PBw7Re/+IXrs8bGRi0wMFB75ZVXvpRn1tfXayKivf322677+/v7a6+99prrOx999JEmItq77777pTxT0zTtjjvu0DZv3nxTz7sVfPcGzzVN8V3xXfG9t/L9RqQ8RS/SmTNnpK6uTibSYWADBw6UBx98UN59990v5RlNTU0iIhIaqh8A/P7778vVq1ctz4yLi5OoqKgv5ZmdnZ2yc+dOaWlpkTFjxtzy590s3Q6eiyi+X0uK74rvfYXvX4mONn2V6owDdO+8807L53feeafrb1+Eurq6JDs7W7797W9LvHHocF1dnQQEBEgInzb8JTzzT3/6k4wZM0ba2trEbrfL7373OxkxYoSUlZXdkuf1lG41z0UU392R4rvi+5fx3NvBd6UUv8I0f/58OXnypBzlY9hvEd17771SVlYmTU1Nsnv3bpk1a5a8/fbbt/y5vZEU371Diu/eoa8a3xV86kUKDw8XEemWIXXhwgXX33pKCxYskP3798sf//hH13E15jPb29ulsbHxS31mQECAxMTEyAMPPCAvvPCC3H///fKrX/3qlj2vp3QreS6i+O6JFN8V37+M594Oviul6EUaNmyYhIeHy1tvveX6zOl0ynvvvSdjxozp0T01TZMFCxbI7373OykqKpJhw4ZZ/v7AAw+Iv7+/5Zkff/yx1NTU9PiZ7qirq0uuXLly2573eelW8FxE8f1GpPiu+N5n+P6lpQH1cbp06ZL2wQcfaB988IEmIlp+fr72wQcfaH/9619v6X1Xr16thYSEaK+//rpWXl6uPfHEE9qwYcO0y5cv9+h5P/rRj7SBAwdqxcXF2vnz513/tba2ur4zd+5cLSoqSisqKtJKS0u1MWPGaGPGjOnxO+bk5Ghvv/22dubMGa28vFzLycnRfHx8tMLCwhs+71bw/XbzXNMU3z/PPRXfFd+9xfebIaUUDfoyz2S8mft2dXVpzz77rHbnnXdqgYGB2oQJE7SPP/64x89z9ywR0bZu3er6zuXLl7V58+Zpd9xxhxYUFKR997vf1c6fP9/jZ2ZmZmpDhw7VAgICtMGDB2sTJkxwCeqNnncr+H67ea5piu+f556K74rv3uL7zZA6OkqRIkWKFCkySMUUFSlSpEiRIoOUUlSkSJEiRYoMUkpRkSJFihQpMkgpRUWKFClSpMggpRQVKVKkSJEig5RSVKRIkSJFigxSSlGRIkWKFCkySClFRYoUKVKkyCClFPsoLV++XBISErw9jL87Unz3Dim+e4f+HvmulOJtptmzZ4uPj4/4+PiIv7+/3HnnnfLoo4/Kli1bpKury9vD+8qS4rt3SPHdO6T43nNSStELNHnyZDl//rxUV1fLwYMH5ZFHHpEf//jHkpqaKh0dHd4e3leWFN+9Q4rv3iHF956RUopeoMDAQAkPD5e7775bRo0aJUuXLpXXX39dDh48KNu2bRMRkcbGRsnKypLBgweLw+GQ8ePHy4cffujxnv/7v/8rjz76qISFhcnAgQPlO9/5jpw4ccL198zMTElNTbX85urVqzJkyBD5zW9+c0ves7eR4rt3SPHdO6T43jNSSrGX0Pjx4+X++++XPXv2iIjIP/3TP0l9fb0cPHhQ3n//fRk1apRMmDBBLl686Pb3ly5dklmzZsnRo0fl2LFjcs8990hKSopcunRJRESysrLk0KFDcv78eddv9u/fL62trfLkk0/e+hfspaT47h1SfPcOKb5/DurxOR6KekSzZs3SnnjiCbd/e/LJJ7VvfOMb2v/8z/9oDodDa2trs/z961//uvbyyy9rmqZpzz33nHb//fd7fE5nZ6cWHBysvfHGG67PRowYob344ouufz/++OPa7Nmze/4yfYgU371Diu/eIcX3npPyFHsRaZomPj4+8uGHH0pzc7MMGjRI7Ha7678zZ87IX/7yF7e/vXDhgvzwhz+Ue+65RwYOHCgOh0Oam5ulpqbG9Z2srCzZunWr6/sHDx6UzMzM2/JuvZkU371Diu/eIcX365OftwegCPTRRx/JsGHDpLm5We666y4pLi7u9p2QkBC3v501a5Z89tln8qtf/UqGDh0qgYGBMmbMGGlvb3d9Z+bMmZKTkyPvvvuulJSUyLBhw+Shhx66RW/Td0jx3Tuk+O4dUny/Piml2EuoqKhI/vSnP8lPfvITiYyMlLq6OvHz85Po6OjP9ft33nlHNmzYICkpKSIicvbsWWloaLB8Z9CgQZKeni5bt26Vd999V37wgx982a/R50jx3Tuk+O4dUny/MSml6AW6cuWK1NXVSWdnp1y4cEEOHTokL7zwgqSmpsrMmTPF19dXxowZI+np6ZKXlyexsbFSW1srv//97+W73/2uJCYmdrvnPffcI7/97W8lMTFRnE6nPPPMM9K/f/9u38vKypLU1FTp7OyUWbNm3Y7X7TWk+O4dUnz3Dim+95C8G9L8+6NZs2ZpIqKJiObn56cNHjxYmzhxorZlyxats7PT9T2n06k9/fTTWkREhObv76997Wtf05566imtpqZG07TuAfATJ05oiYmJms1m0+655x7ttdde04YOHar98pe/tDy/q6tLGzp0qJaSknI7XrfXkOK7d0jx3Tuk+N5z8tE0TfOeSlZ0u6m5uVnuvvtu2bp1q2RkZHh7OH83pPjuHVJ89w71Zb4r+PTvhLq6uqShoUFeeuklCQkJkbS0NG8P6e+CFN+9Q4rv3qGvAt+VUvw7oZqaGhk2bJhERkbKtm3bxM9PTf3tIMV375Diu3foq8B3BZ8qUqRIkSJFBqnifUWKFClSpMggpRQVKVKkSJEig5RSVKRIkSJFigxSSlGRIkWKFCkySClFRYoUKVKkyCClFBUpUqRIkSKDlFJUpEiRIkWKDFJKUZEiRYoUKTLo/wfshUrWfbt8qAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 450x278.116 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = env.plot_SM(SM_start, colors=colors, N_SM_show=4, subplotpars=subplotpars_learn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "0facdaf5-5d3f-4f4e-9948-6e4c24228d8b",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Net(N_pre, N_SMs, N_SM_time)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "e420b413-e47b-4956-9c45-58baf02db2e2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([233, 128, 31])"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.conv.weight.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "caf0fb1a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor(0., device='mps:0'), tensor(1., device='mps:0'))"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b_true = env.get_b(seed=seed).to(device)\n",
    "# generate the corresponding raster plot = what is observed\n",
    "a = env.draw_a(b_true, SM_true.to(device), seed=seed+1)\n",
    "a.min(), a.max()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "2699253a",
   "metadata": {},
   "source": [
    "## Supervised Learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "301fd4fa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2023-09-27_SL_THC_DEBUG=1.pth\n",
      "2023-09-27_US_lr=1.00e-03_THC_DEBUG=1.pth\n",
      "2023-09-27_US_lr=1.00e-05_THC_DEBUG=1.pth\n",
      "2023-09-27_US_lr=1.58e-04_THC_DEBUG=1.pth\n",
      "2023-09-27_US_lr=2.51e-05_THC_DEBUG=1.pth\n",
      "2023-09-27_US_lr=3.98e-04_THC_DEBUG=1.pth\n",
      "2023-09-27_US_lr=6.31e-05_THC_DEBUG=1.pth\n",
      "2023-09-27_scan_E_SM.json\n",
      "2023-09-27_scan_N_SM_time.json\n",
      "2023-09-27_scan_N_SMs.json\n",
      "2023-09-27_scan_N_pre.json\n",
      "2023-09-27_scan_p_A.json\n",
      "2023-09-27_scan_p_B.json\n",
      "2023-09-27_scan_p_SM.json\n",
      "2023-09-27_scan_tau_decay.json\n",
      "2023-09-27_scan_tau_rise.json\n"
     ]
    }
   ],
   "source": [
    "%ls data_cache"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "32cb2463-191c-4a8b-a385-d4eeb2db3715",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "model = learn_model(cachepath(data_cache, datetag + '_SL', DEBUG), env, SM_start, SM_true, do_supervised=True, N_learnsteps=N_learnsteps)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "eae8531f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(233, 128, 31)\n",
      "-0.061568767 0.08482143 0.11774757\n",
      "tensor(-8.854) tensor(0.000) tensor(10.350)\n"
     ]
    }
   ],
   "source": [
    "SM_hat = model.conv.weight.detach().cpu().numpy()\n",
    "print(SM_hat.shape)\n",
    "print(SM_hat.min(), SM_hat.mean(), SM_hat.max())\n",
    "print(SM_true.min(), SM_true.mean(), SM_true.max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "7575e6ba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 450x278.116 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = env.plot_SM(SM_true, colors=colors, N_SM_show=4, subplotpars=subplotpars_learn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "7c4fae86",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 450x278.116 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = env.plot_SM(SM_hat, colors=colors, N_SM_show=4, subplotpars=subplotpars_learn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "c4cc2f98",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(128, (128, 128), array([ 0, 32, 64, 96]), [1, 32, 64, 96, 128], 0.0071105068)"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Xcorr = (SM_hat[:, None, :, :] * SM_true.numpy()[:, :, None, :]).mean(axis=(0, -1))\n",
    "env.opt.N_pre, Xcorr.shape, np.arange(0, env.opt.N_pre, env.opt.N_pre//4), [1] + (np.arange(0, env.opt.N_pre, env.opt.N_pre//4) + env.opt.N_pre//4).tolist(), Xcorr.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "293cb128",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 325x325 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "subplotpars_scan = matplotlib.figure.SubplotParams(left=0.125, right=.95, bottom=0.25, top=.975)\n",
    "fig, ax = plt.subplots(figsize=(opt_scan.fig_width, opt_scan.fig_width), subplotpars=subplotpars_scan)\n",
    "ax.imshow(Xcorr, interpolation='nearest', vmin=0)\n",
    "\n",
    "ax.set_ylabel('True')\n",
    "ax.set_xlabel('Learned')\n",
    "\n",
    "ticks = [.5] + (np.arange(0, env.opt.N_pre, env.opt.N_pre//4) + env.opt.N_pre//4 - .5).tolist()\n",
    "ticklabels = [1] + (np.arange(0, env.opt.N_pre, env.opt.N_pre//4) + env.opt.N_pre//4).astype(int).tolist()\n",
    "ax.set_xticks(ticks)\n",
    "ax.set_xticklabels(ticklabels, rotation=60)\n",
    "ax.set_yticks(ticks)\n",
    "ax.set_yticklabels(ticklabels, rotation=60)\n",
    "\n",
    "if not figpath is None: printfig(fig, 'THC_xcorr-supervised')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.5"
  },
  "toc-autonumbering": true,
  "vscode": {
   "interpreter": {
    "hash": "1a1af0ee75eeea9e2e1ee996c87e7a2b11a0bebd85af04bb136d915cefc0abce"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}