{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torchaudio\n",
    "import torchaudio.transforms as T\n",
    "from typing import List\n",
    "import pyloudnorm as pyln\n",
    "import numpy as np\n",
    "\n",
    "def compute_band_energy(waveform: torch.Tensor, \n",
    "                       sample_rate: int,\n",
    "                       n_fft: int = 4096) -> float:\n",
    "    bands = {\n",
    "        'bass': (20, 250),\n",
    "        'mid': (250, 2500),\n",
    "        'high': (2500, 20000)\n",
    "    }\n",
    "\n",
    "    # Convert to mono if stereo\n",
    "    if waveform.shape[0] > 1:\n",
    "        waveform = torch.mean(waveform, dim=0, keepdim=True)\n",
    "    \n",
    "    # Split into frames\n",
    "    frame_length = n_fft\n",
    "    hop_length = n_fft // 2  # 50% overlap\n",
    "    frames = waveform.unfold(1, frame_length, hop_length)\n",
    "    # Compute FFT for each frame\n",
    "    spectrum = torch.fft.rfft(frames.squeeze(0))  # [num_frames, n_fft//2 + 1]\n",
    "    freqs = torch.fft.rfftfreq(n_fft, d=1/sample_rate)  # [n_fft//2 + 1]\n",
    "    \n",
    "    # Compute magnitudes for each frame\n",
    "    magnitudes =torch.abs(spectrum)  # [num_frames, n_fft//2 + 1]\n",
    "    \n",
    "    # Compute centroid for each frame\n",
    "    numerator = torch.sum(freqs.view(1, -1) * magnitudes, dim=1)  # Sum over frequencies for each frame\n",
    "    denominator = torch.sum(magnitudes, dim=1)\n",
    "    \n",
    "    results = []\n",
    "\n",
    "    # Compute mean centroid across all frames\n",
    "    centroid = torch.mean(numerator / (denominator + 1e-8))\n",
    "    results.append(float(centroid))\n",
    "    \n",
    "    for band_name, (low_freq, high_freq) in bands.items():\n",
    "        # Create frequency mask\n",
    "        mask = (freqs >= low_freq) & (freqs <= high_freq)\n",
    "                \n",
    "        # They should now have the same size\n",
    "        band_energy = torch.mean(magnitudes * mask)\n",
    "        results.append(float(band_energy))\n",
    "\n",
    "    return results\n",
    "\n",
    "def calculate_stereo_width(waveform):\n",
    "    # Split into left and right channels\n",
    "    left = waveform[0]\n",
    "    right = waveform[1]\n",
    "  \n",
    "    # Compute mid/side representation\n",
    "    mid = (left + right) / 2\n",
    "    side = (left - right) / 2\n",
    "    \n",
    "    # Compute RMS energy of mid and side channels\n",
    "    mid_energy = torch.sqrt(torch.mean(mid ** 2))\n",
    "    side_energy = torch.sqrt(torch.mean(side ** 2))\n",
    "    \n",
    "    # Compute stereo width based on mid/side ratio\n",
    "    # Normalize to range 0-1 using sigmoid-like function\n",
    "    width_ratio = (side_energy / (mid_energy + 1e-8)).item()\n",
    "    stereo_width = 2 * (1 / (1 + np.exp(-width_ratio)) - 0.5)\n",
    "\n",
    "    return stereo_width.item()\n",
    "\n",
    "def calculate_crest_factor(signal, window_size):\n",
    "    \"\"\"\n",
    "    Calculate RMS values for windows of audio data using PyTorch.\n",
    "    \n",
    "    Parameters:\n",
    "    signal (torch.Tensor): Audio signal tensor\n",
    "    window_size (int): Size of the window for RMS calculation\n",
    "    \n",
    "    Returns:\n",
    "    torch.Tensor: Tensor of RMS values\n",
    "    \"\"\"\n",
    "    # Ensure input is a tensor\n",
    "    if not isinstance(signal, torch.Tensor):\n",
    "        signal = torch.tensor(signal, dtype=torch.float32)\n",
    "    \n",
    "    # convert to mono if stereo\n",
    "    if signal.shape[0] > 1:\n",
    "        signal = signal.mean(dim=0)\n",
    "\n",
    "    # Calculate pad size\n",
    "    pad_size = window_size - (len(signal) % window_size)\n",
    "    if pad_size < window_size:\n",
    "        # Use constant padding (default value is 0)\n",
    "        padded_signal = torch.nn.functional.pad(signal, (0, pad_size))\n",
    "    else:\n",
    "        padded_signal = signal\n",
    "    \n",
    "    # Reshape signal into windows using unfold\n",
    "    # unfold(dimension, size, step) creates overlapping windows\n",
    "    # here we use step=size to create non-overlapping windows\n",
    "    windows = padded_signal.unfold(0, window_size, window_size)\n",
    "    \n",
    "    # Calculate RMS for each window\n",
    "    # torch.mean along dim=1 averages across the window\n",
    "    # keepdim=False reduces the dimension\n",
    "    window_rms = torch.sqrt(torch.mean(windows**2, dim=1))\n",
    "    rms = torch.mean(window_rms).item()\n",
    "\n",
    "    # get the peak value\n",
    "    peak = torch.max(torch.abs(signal)).item()\n",
    "\n",
    "    # compute crest factor\n",
    "    crest_factor = peak / (rms + 1e-8)\n",
    "\n",
    "    return np.log(crest_factor + 1e-8)\n",
    "\n",
    "import torch\n",
    "import torchaudio\n",
    "\n",
    "def measure_silence_percentage(\n",
    "    waveform: torch.Tensor,\n",
    "    sample_rate: int,\n",
    "    silence_threshold_db: float = -60,\n",
    "    window_size_ms: int = 100\n",
    ") -> float:\n",
    "    \"\"\"\n",
    "    Measures the percentage of audio that could be considered silent.\n",
    "    \n",
    "    Args:\n",
    "        file_path: Path to audio file\n",
    "        silence_threshold_db: RMS threshold in dB below which audio is considered silent\n",
    "        window_size_ms: Size of analysis window in milliseconds\n",
    "    \n",
    "    Returns:\n",
    "        Percentage (0-100) of audio that is below the silence threshold\n",
    "    \"\"\"\n",
    "    \n",
    "    # Convert to mono if stereo\n",
    "    if waveform.shape[0] > 1:\n",
    "        waveform = torch.mean(waveform, dim=0, keepdim=True)\n",
    "    \n",
    "    # Calculate window size in samples\n",
    "    window_size = int(sample_rate * window_size_ms / 1000)\n",
    "    \n",
    "    # Unfold the waveform into windows\n",
    "    windows = waveform.unfold(1, window_size, window_size)\n",
    "    \n",
    "    # Calculate RMS for each window\n",
    "    rms = torch.sqrt(torch.mean(windows ** 2, dim=2))\n",
    "    db = 20 * torch.log10(rms + 1e-10)\n",
    "    \n",
    "    # Calculate percentage of windows below threshold\n",
    "    silence_percentage = 100 * torch.mean((db < silence_threshold_db).float()).item()\n",
    "    \n",
    "    return silence_percentage\n",
    "\n",
    "\n",
    "def measure_spectral_flatness(audio_tensor):\n",
    "    \"\"\"\n",
    "    Calculate spectral flatness (Wiener entropy) of the signal.\n",
    "    Returns value between 0 (pure tone) and 1 (white noise).\n",
    "    \n",
    "    Parameters:\n",
    "    audio_tensor: Input audio tensor of shape [..., samples]\n",
    "    \n",
    "    Returns:\n",
    "    torch.Tensor: Spectral flatness value between 0 and 1\n",
    "    \"\"\"\n",
    "    # Get spectrum magnitude\n",
    "    audio_tensor = audio_tensor.mean(dim=0)\n",
    "    spectrum = torch.abs(torch.fft.rfft(audio_tensor, dim=-1))\n",
    "    \n",
    "    # Add small epsilon to avoid log(0)\n",
    "    epsilon = 1e-8\n",
    "    spectrum = spectrum + epsilon\n",
    "    \n",
    "    # Calculate geometric mean and arithmetic mean\n",
    "    log_spectrum = torch.log(spectrum)\n",
    "    geometric_mean = torch.exp(torch.mean(log_spectrum, dim=-1))\n",
    "    arithmetic_mean = torch.mean(spectrum, dim=-1)\n",
    "    \n",
    "    # Compute flatness\n",
    "    flatness = geometric_mean / (arithmetic_mean + 1e-8)\n",
    "    \n",
    "    return flatness.item()\n",
    "\n",
    "class AudioProductionFeatures(torch.utils.data.Dataset):\n",
    "    def __init__(self, filepaths: List[str], sample_rate: int = 48000, target_loudness: float = -16.0):\n",
    "        self.filepaths = filepaths\n",
    "        self.sample_rate = sample_rate\n",
    "        self.target_loudness = target_loudness\n",
    "        self.features = {\n",
    "            'bass': [],\n",
    "            'mid': [],\n",
    "            'high': []\n",
    "        }\n",
    "    \n",
    "        self.meter = pyln.Meter(self.sample_rate)\n",
    "        print(f\"Loaded {len(self.filepaths)} files\")\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.filepaths)\n",
    "    \n",
    "    def __getitem__(self, idx):\n",
    "        filepath = self.filepaths[idx]\n",
    "        x, sr = torchaudio.load(filepath)\n",
    "\n",
    "        # for speed, only use first 120s\n",
    "        x = x[:, :int(60 * sr)]\n",
    "\n",
    "        # resample if necessary\n",
    "        if sr != self.sample_rate:\n",
    "            x = torchaudio.functional.resample(x, sr, self.sample_rate)\n",
    "\n",
    "        # loudness normalization\n",
    "        loudness = self.meter.integrated_loudness(x.permute(1, 0).numpy())\n",
    "        # check if loudness is -inf\n",
    "        if loudness == -np.inf:\n",
    "            loudness = -80.0\n",
    "        loudness_diff = self.target_loudness - loudness\n",
    "        # limit the loudness difference to +/- 20 db\n",
    "        loudness_diff = np.clip(loudness_diff, -20, 20)\n",
    "        x *= 10 ** (loudness_diff / 20.0)\n",
    "        # this can cause problems\n",
    "\n",
    "        # compute features\n",
    "        with torch.no_grad():\n",
    "            results_spectral = compute_band_energy(x, self.sample_rate)\n",
    "            results_crest_factor = calculate_crest_factor(x, 1024)\n",
    "            results_stereo_width = calculate_stereo_width(x)\n",
    "            results_spectral_flatness = measure_spectral_flatness(x)\n",
    "            results_silence_percentage = measure_silence_percentage(x, self.sample_rate)\n",
    "\n",
    "        results = results_spectral + [results_crest_factor, results_stereo_width, results_spectral_flatness, results_silence_percentage, loudness]\n",
    "\n",
    "        return filepath, torch.tensor(results)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "179999\n",
      "Loaded 10000 files\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10000/10000 [01:25<00:00, 117.53it/s]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>filepath</th>\n",
       "      <th>spectral_centroid</th>\n",
       "      <th>bass</th>\n",
       "      <th>mid</th>\n",
       "      <th>high</th>\n",
       "      <th>crest_factor</th>\n",
       "      <th>stereo_width</th>\n",
       "      <th>spectral_flatness</th>\n",
       "      <th>silence_percentage</th>\n",
       "      <th>loudness</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>/app/suno/data/audio_2ch_48khz_lg/train/genius...</td>\n",
       "      <td>3536.962891</td>\n",
       "      <td>0.349990</td>\n",
       "      <td>0.776892</td>\n",
       "      <td>1.030472</td>\n",
       "      <td>1.253951</td>\n",
       "      <td>0.099555</td>\n",
       "      <td>0.192333</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-15.667814</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>/app/suno/data/audio_2ch_48khz_lg/train/genius...</td>\n",
       "      <td>4095.244141</td>\n",
       "      <td>0.265834</td>\n",
       "      <td>0.788508</td>\n",
       "      <td>1.033199</td>\n",
       "      <td>1.850788</td>\n",
       "      <td>0.129069</td>\n",
       "      <td>0.164842</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-16.258949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>/app/suno/data/audio_2ch_48khz_lg/train/genius...</td>\n",
       "      <td>4292.973633</td>\n",
       "      <td>0.281722</td>\n",
       "      <td>0.741110</td>\n",
       "      <td>1.172707</td>\n",
       "      <td>1.288747</td>\n",
       "      <td>0.151085</td>\n",
       "      <td>0.236274</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-15.384681</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>/app/suno/data/audio_2ch_48khz_lg/train/genius...</td>\n",
       "      <td>3031.575439</td>\n",
       "      <td>0.307406</td>\n",
       "      <td>0.947427</td>\n",
       "      <td>0.844501</td>\n",
       "      <td>1.114733</td>\n",
       "      <td>0.224983</td>\n",
       "      <td>0.139384</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-15.840470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>/app/suno/data/audio_2ch_48khz_lg/train/genius...</td>\n",
       "      <td>3406.133545</td>\n",
       "      <td>0.381965</td>\n",
       "      <td>0.794475</td>\n",
       "      <td>0.918959</td>\n",
       "      <td>1.124280</td>\n",
       "      <td>0.276420</td>\n",
       "      <td>0.192652</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-15.802269</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            filepath  spectral_centroid  \\\n",
       "0  /app/suno/data/audio_2ch_48khz_lg/train/genius...        3536.962891   \n",
       "1  /app/suno/data/audio_2ch_48khz_lg/train/genius...        4095.244141   \n",
       "2  /app/suno/data/audio_2ch_48khz_lg/train/genius...        4292.973633   \n",
       "3  /app/suno/data/audio_2ch_48khz_lg/train/genius...        3031.575439   \n",
       "4  /app/suno/data/audio_2ch_48khz_lg/train/genius...        3406.133545   \n",
       "\n",
       "       bass       mid      high  crest_factor  stereo_width  \\\n",
       "0  0.349990  0.776892  1.030472      1.253951      0.099555   \n",
       "1  0.265834  0.788508  1.033199      1.850788      0.129069   \n",
       "2  0.281722  0.741110  1.172707      1.288747      0.151085   \n",
       "3  0.307406  0.947427  0.844501      1.114733      0.224983   \n",
       "4  0.381965  0.794475  0.918959      1.124280      0.276420   \n",
       "\n",
       "   spectral_flatness  silence_percentage   loudness  \n",
       "0           0.192333                 0.0 -15.667814  \n",
       "1           0.164842                 0.0 -16.258949  \n",
       "2           0.236274                 0.0 -15.384681  \n",
       "3           0.139384                 0.0 -15.840470  \n",
       "4           0.192652                 0.0 -15.802269  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import glob\n",
    "from tqdm import tqdm\n",
    "import pandas as pd\n",
    "#filepaths = glob.glob(\"/home/christian/audio/reference-audio-wav/*.wav\")\n",
    "filepaths = glob.glob(\"/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/*.wav\")\n",
    "print(len(filepaths))\n",
    "filepaths = filepaths[:10000]\n",
    "\n",
    "dataset = AudioProductionFeatures(filepaths, sample_rate=48000, target_loudness=-16.0)\n",
    "dataloader = torch.utils.data.DataLoader(dataset, batch_size=1, shuffle=False, num_workers=96)\n",
    "\n",
    "data = []\n",
    "\n",
    "for filepath, results in tqdm(dataloader):\n",
    "    result_list = results[0].tolist()\n",
    "    full_result_list = [filepath[0]] + result_list\n",
    "    data.append(full_result_list)\n",
    "\n",
    "data = pd.DataFrame(data, columns=['filepath', 'spectral_centroid', 'bass', 'mid', 'high', 'crest_factor', 'stereo_width', 'spectral_flatness', 'silence_percentage', 'loudness'])\n",
    "\n",
    "data.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>spectral_centroid</th>\n",
       "      <th>bass</th>\n",
       "      <th>mid</th>\n",
       "      <th>high</th>\n",
       "      <th>crest_factor</th>\n",
       "      <th>stereo_width</th>\n",
       "      <th>spectral_flatness</th>\n",
       "      <th>silence_percentage</th>\n",
       "      <th>loudness</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>3281.356469</td>\n",
       "      <td>0.333578</td>\n",
       "      <td>0.660989</td>\n",
       "      <td>0.693207</td>\n",
       "      <td>1.595248</td>\n",
       "      <td>0.137874</td>\n",
       "      <td>0.152950</td>\n",
       "      <td>0.278611</td>\n",
       "      <td>-16.231639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>713.305302</td>\n",
       "      <td>0.097143</td>\n",
       "      <td>0.132886</td>\n",
       "      <td>0.273209</td>\n",
       "      <td>0.477831</td>\n",
       "      <td>0.077062</td>\n",
       "      <td>0.059294</td>\n",
       "      <td>2.371171</td>\n",
       "      <td>1.385460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-18.420681</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.006799</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-80.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2788.736816</td>\n",
       "      <td>0.275140</td>\n",
       "      <td>0.574327</td>\n",
       "      <td>0.497982</td>\n",
       "      <td>1.324464</td>\n",
       "      <td>0.083183</td>\n",
       "      <td>0.112794</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-16.358351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3253.377808</td>\n",
       "      <td>0.338676</td>\n",
       "      <td>0.664741</td>\n",
       "      <td>0.680789</td>\n",
       "      <td>1.542473</td>\n",
       "      <td>0.130261</td>\n",
       "      <td>0.152203</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-15.990255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>3736.420288</td>\n",
       "      <td>0.399001</td>\n",
       "      <td>0.750537</td>\n",
       "      <td>0.872532</td>\n",
       "      <td>1.816164</td>\n",
       "      <td>0.185326</td>\n",
       "      <td>0.190838</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-15.740542</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>8253.011719</td>\n",
       "      <td>0.748593</td>\n",
       "      <td>1.254508</td>\n",
       "      <td>2.519476</td>\n",
       "      <td>4.589586</td>\n",
       "      <td>0.724192</td>\n",
       "      <td>0.499998</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>-13.419356</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       spectral_centroid          bass           mid          high  \\\n",
       "count       10000.000000  10000.000000  10000.000000  10000.000000   \n",
       "mean         3281.356469      0.333578      0.660989      0.693207   \n",
       "std           713.305302      0.097143      0.132886      0.273209   \n",
       "min             0.000000      0.000000      0.000000      0.000000   \n",
       "25%          2788.736816      0.275140      0.574327      0.497982   \n",
       "50%          3253.377808      0.338676      0.664741      0.680789   \n",
       "75%          3736.420288      0.399001      0.750537      0.872532   \n",
       "max          8253.011719      0.748593      1.254508      2.519476   \n",
       "\n",
       "       crest_factor  stereo_width  spectral_flatness  silence_percentage  \\\n",
       "count  10000.000000  10000.000000       10000.000000        10000.000000   \n",
       "mean       1.595248      0.137874           0.152950            0.278611   \n",
       "std        0.477831      0.077062           0.059294            2.371171   \n",
       "min      -18.420681      0.000000           0.006799            0.000000   \n",
       "25%        1.324464      0.083183           0.112794            0.000000   \n",
       "50%        1.542473      0.130261           0.152203            0.000000   \n",
       "75%        1.816164      0.185326           0.190838            0.000000   \n",
       "max        4.589586      0.724192           0.499998          100.000000   \n",
       "\n",
       "           loudness  \n",
       "count  10000.000000  \n",
       "mean     -16.231639  \n",
       "std        1.385460  \n",
       "min      -80.000000  \n",
       "25%      -16.358351  \n",
       "50%      -15.990255  \n",
       "75%      -15.740542  \n",
       "max      -13.419356  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "crest factor\n",
      "4.589585640448382\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/03d40105-47d6-470b-802b-2f9192cca777.wav\n",
      "-18.420680743952364\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/8e9dd372-b4a0-4ab8-8ee9-b19b310ffa05.wav\n",
      "\n",
      "stereo width\n",
      "0.7241916338312879\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/c33221f5-9943-453d-ac70-c57b033944c5.wav\n",
      "0.0\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/bc0a34a2-e8a9-4b0b-b519-ec14534aad12.wav\n",
      "\n",
      "spectral flatness\n",
      "0.4999980032444\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/8e9dd372-b4a0-4ab8-8ee9-b19b310ffa05.wav\n",
      "0.006799295544624329\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/d225caeb-e346-470e-9028-e9da32cd92d8.wav\n",
      "\n",
      "silence percentage\n",
      "100.0\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/8e9dd372-b4a0-4ab8-8ee9-b19b310ffa05.wav\n",
      "0.0\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/0ae8e173-725e-4f93-bdf1-8bdc6ac935b8.wav\n",
      "\n",
      "spectral centroid\n",
      "8253.01171875\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/bbe4d289-b61a-4bf1-9c78-4c7d5252a6cc.wav\n",
      "0.0\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/8e9dd372-b4a0-4ab8-8ee9-b19b310ffa05.wav\n"
     ]
    }
   ],
   "source": [
    "# find the filepath with the highest crest factor\n",
    "print(\"crest factor\")\n",
    "print(data['crest_factor'].max())\n",
    "filepath = data[data['crest_factor'] == data['crest_factor'].max()]['filepath'].values[0]\n",
    "print(filepath)\n",
    "\n",
    "# find the filepath with the lowest crest factor\n",
    "print(data['crest_factor'].min())\n",
    "filepath = data[data['crest_factor'] == data['crest_factor'].min()]['filepath'].values[0]\n",
    "print(filepath)\n",
    "\n",
    "print(\"\\nstereo width\")\n",
    "print(data['stereo_width'].max())\n",
    "filepath = data[data['stereo_width'] == data['stereo_width'].max()]['filepath'].values[0]\n",
    "print(filepath)\n",
    "\n",
    "# find the filepath with the lowest stereo width\n",
    "print(data['stereo_width'].min())\n",
    "filepath = data[data['stereo_width'] == data['stereo_width'].min()]['filepath'].values[0]\n",
    "print(filepath)\n",
    "\n",
    "print(\"\\nspectral flatness\")\n",
    "print(data['spectral_flatness'].max())\n",
    "filepath = data[data['spectral_flatness'] == data['spectral_flatness'].max()]['filepath'].values[0]\n",
    "print(filepath)\n",
    "\n",
    "# find the filepath with the lowest spectral flatness\n",
    "print(data['spectral_flatness'].min())\n",
    "filepath = data[data['spectral_flatness'] == data['spectral_flatness'].min()]['filepath'].values[0]\n",
    "print(filepath)\n",
    "\n",
    "print(\"\\nsilence percentage\")\n",
    "print(data['silence_percentage'].max())\n",
    "filepath = data[data['silence_percentage'] == data['silence_percentage'].max()]['filepath'].values[0]\n",
    "print(filepath)\n",
    "\n",
    "# find the filepath with the lowest silence percentage\n",
    "print(data['silence_percentage'].min())\n",
    "filepath = data[data['silence_percentage'] == data['silence_percentage'].min()]['filepath'].values[0]\n",
    "print(filepath)\n",
    "\n",
    "print(\"\\nspectral centroid\")\n",
    "print(data['spectral_centroid'].max())\n",
    "filepath = data[data['spectral_centroid'] == data['spectral_centroid'].max()]['filepath'].values[0]\n",
    "print(filepath)\n",
    "\n",
    "# find the filepath with the lowest spectral centroid\n",
    "print(data['spectral_centroid'].min())\n",
    "filepath = data[data['spectral_centroid'] == data['spectral_centroid'].min()]['filepath'].values[0]\n",
    "print(filepath)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# create a new crestor factor column that is log scaled\n",
    "import matplotlib.pyplot as plt\n",
    "plt.hist(data['crest_factor'], bins=100)\n",
    "# put the values on top of the bars in the histogram\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "bass mean: 166.0 (55.3)\n",
      "bass min: -3.0 max: 4.4\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mid mean: 114.8 (25.1)\n",
      "mid min: -4.6 max: 4.2\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "high mean: 45.7 (15.5)\n",
      "high min: -3.0 max: 4.7\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAioAAAGdCAYAAAA8F1jjAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy80BEi2AAAACXBIWXMAAA9hAAAPYQGoP6dpAAAjl0lEQVR4nO3de2xUdf7/8Vdb6HDrTC3SDl1abl4AuWi4lFH0y6VLKV2UpRpwWahIYCEDEboi1iCou25ZYhZWguBmXdCEBi+7QEDlYpWyhnKrS0CQRgikYJmCkHagG1po5/fH98fkO1C0U2Y4n06fj+QknHM+c+b9EWlf+ZzP+Zwon8/nEwAAgIGirS4AAADgdggqAADAWAQVAABgLIIKAAAwFkEFAAAYi6ACAACMRVABAADGIqgAAABjtbK6gKaor69XeXm54uLiFBUVZXU5AACgEXw+ny5fvqzk5GRFRzdurKRZBpXy8nKlpKRYXQYAAGiCM2fOqEuXLo1q2yyDSlxcnKT/7ajdbre4GgAA0Bher1cpKSn+3+ON0SyDyo3bPXa7naACAEAzE8y0DSbTAgAAYxFUAACAsQgqAADAWAQVAABgLIIKAAAwFkEFAAAYi6ACAACMRVABAADGIqgAAABjEVQAAICxCCoAAMBYBBUAAGAsggoAADAWQQUAABirldUFAAivbi9/esux00uzLKgEAILHiAoAADAWQQUAABiLoAIAAIzFHBUggjQ0HwUAmjNGVAAAgLEIKgAAwFgEFQAAYCyCCgAAMBZBBQAAGIunfgBD3fwED6vJAmiJGFEBAADGIqgAAABjcesHQKPwckMAVmBEBQAAGIugAgAAjEVQAQAAxiKoAAAAYxFUAACAsYIKKqtXr1b//v1lt9tlt9vlcrn0+eef+89fvXpVbrdbHTt2VIcOHZSdna2KioqAa5SVlSkrK0vt2rVTYmKiFixYoOvXr4emNwAAIKIEFVS6dOmipUuXqqSkRAcPHtTIkSP11FNP6ejRo5Kk+fPna8uWLfr4449VVFSk8vJyTZgwwf/5uro6ZWVlqba2Vnv27NH777+vdevWafHixaHtFQAAiAhBraMybty4gP0333xTq1ev1t69e9WlSxe99957Kigo0MiRIyVJa9euVe/evbV3714NHTpUO3bs0LFjx/TFF18oKSlJDz/8sP7whz9o4cKFeu211xQbGxu6ngEAgGavyXNU6urqtGHDBlVXV8vlcqmkpETXrl1Tenq6v02vXr2Umpqq4uJiSVJxcbH69eunpKQkf5uMjAx5vV7/qExDampq5PV6AzYAABD5gg4qR44cUYcOHWSz2TRr1ixt3LhRffr0kcfjUWxsrOLj4wPaJyUlyePxSJI8Hk9ASLlx/sa528nPz5fD4fBvKSkpwZYNAACaoaCX0H/wwQd16NAhVVVV6ZNPPlFOTo6KiorCUZtfXl6ecnNz/fter5ewAtwB3swMoLkIOqjExsbqvvvukyQNHDhQBw4c0F//+ldNnDhRtbW1qqysDBhVqaiokNPplCQ5nU7t378/4Ho3ngq60aYhNptNNpst2FIBNBLv8QFgqjteR6W+vl41NTUaOHCgWrdurcLCQv+50tJSlZWVyeVySZJcLpeOHDmi8+fP+9vs3LlTdrtdffr0udNSAABAhAlqRCUvL0+ZmZlKTU3V5cuXVVBQoF27dmn79u1yOByaPn26cnNzlZCQILvdrrlz58rlcmno0KGSpNGjR6tPnz6aMmWKli1bJo/Ho0WLFsntdjNiAgAAbhFUUDl//rymTp2qc+fOyeFwqH///tq+fbt++ctfSpKWL1+u6OhoZWdnq6amRhkZGXrnnXf8n4+JidHWrVs1e/ZsuVwutW/fXjk5OXrjjTdC2ysAABARonw+n8/qIoLl9XrlcDhUVVUlu91udTlAWDQ0b+RmN88jacxnGqsx12YeC4BgNOX3N+/6AQAAxgr6qR8A5gjlCAoAmIigAqBBhCAAJuDWDwAAMBZBBQAAGIugAgAAjEVQAQAAxiKoAAAAYxFUAACAsQgqAADAWAQVAABgLBZ8A0Ls5oXSeB8OADQdIyoAAMBYBBUAAGAsggoAADAWc1SAMGvo5X7MWwGAxiGoAAbgTcUA0DBu/QAAAGMRVAAAgLEIKgAAwFgEFQAAYCwm0wK4q3gKCkAwGFEBAADGIqgAAABjEVQAAICxCCoAAMBYBBUAAGAsnvoBLMCS+QDQOIyoAAAAYxFUAACAsQgqAADAWAQVAABgLIIKAAAwFkEFAAAYi6ACAACMRVABAADGIqgAAABjEVQAAICxCCoAAMBYvOsHCALv6AGAu4sRFQAAYCyCCgAAMBa3fgA02c23wk4vzbKoEgCRihEVAABgrKCCSn5+vgYPHqy4uDglJiZq/PjxKi0tDWgzfPhwRUVFBWyzZs0KaFNWVqasrCy1a9dOiYmJWrBgga5fv37nvQEAABElqFs/RUVFcrvdGjx4sK5fv65XXnlFo0eP1rFjx9S+fXt/uxkzZuiNN97w77dr187/57q6OmVlZcnpdGrPnj06d+6cpk6dqtatW+tPf/pTCLoEAAAiRVBBZdu2bQH769atU2JiokpKSvTEE0/4j7dr105Op7PBa+zYsUPHjh3TF198oaSkJD388MP6wx/+oIULF+q1115TbGxsE7oBAAAi0R3NUamqqpIkJSQkBBxfv3697r33XvXt21d5eXn673//6z9XXFysfv36KSkpyX8sIyNDXq9XR48ebfB7ampq5PV6AzYAABD5mvzUT319vebNm6fHHntMffv29R//zW9+o65duyo5OVmHDx/WwoULVVpaqn/961+SJI/HExBSJPn3PR5Pg9+Vn5+v119/vamlAgCAZqrJQcXtduvbb7/V119/HXB85syZ/j/369dPnTt31qhRo3Ty5En17NmzSd+Vl5en3Nxc/77X61VKSkrTCgcAAM1Gk279zJkzR1u3btVXX32lLl26/GTbtLQ0SdKJEyckSU6nUxUVFQFtbuzfbl6LzWaT3W4P2AAAQOQLKqj4fD7NmTNHGzdu1Jdffqnu3bv/7GcOHTokSercubMkyeVy6ciRIzp//ry/zc6dO2W329WnT59gygEAABEuqFs/brdbBQUF2rx5s+Li4vxzShwOh9q2bauTJ0+qoKBAY8eOVceOHXX48GHNnz9fTzzxhPr37y9JGj16tPr06aMpU6Zo2bJl8ng8WrRokdxut2w2W+h7CAAAmq2gRlRWr16tqqoqDR8+XJ07d/ZvH374oSQpNjZWX3zxhUaPHq1evXrp97//vbKzs7Vlyxb/NWJiYrR161bFxMTI5XLpt7/9raZOnRqw7goAAIAU5IiKz+f7yfMpKSkqKir62et07dpVn332WTBfDQAAWiDe9QMAAIxFUAEAAMYiqAAAAGMRVAAAgLEIKgAAwFgEFQAAYCyCCgAAMFaTX0oIADfr9vKntxw7vTTLgkoARApGVAAAgLEIKgAAwFgEFQAAYCyCCgAAMBZBBQAAGIugAgAAjEVQAQAAxiKoAAAAYxFUAACAsQgqAADAWAQVAABgLIIKAAAwFkEFAAAYi6ACAACMRVABAADGIqgAAABjEVQAAICxWlldAIDI1u3lT60uAUAzRlABYLmbw8zppVkWVQLANNz6AQAAxmJEBbgNblkAgPUYUQEAAMYiqAAAAGMRVAAAgLEIKgAAwFgEFQAAYCyCCgAAMBZBBQAAGIugAgAAjMWCb8D/xwJvAGAeRlQAAICxCCoAAMBYBBUAAGAsggoAADAWk2kBNAs3T3Y+vTTLokoA3E2MqAAAAGMFFVTy8/M1ePBgxcXFKTExUePHj1dpaWlAm6tXr8rtdqtjx47q0KGDsrOzVVFREdCmrKxMWVlZateunRITE7VgwQJdv379znsDAAAiSlBBpaioSG63W3v37tXOnTt17do1jR49WtXV1f428+fP15YtW/Txxx+rqKhI5eXlmjBhgv98XV2dsrKyVFtbqz179uj999/XunXrtHjx4tD1CgAARIQon8/na+qHL1y4oMTERBUVFemJJ55QVVWVOnXqpIKCAj399NOSpOPHj6t3794qLi7W0KFD9fnnn+tXv/qVysvLlZSUJElas2aNFi5cqAsXLig2NvZnv9fr9crhcKiqqkp2u72p5QMBWPDNHA3NP2GOCtD8NeX39x3NUamqqpIkJSQkSJJKSkp07do1paen+9v06tVLqampKi4uliQVFxerX79+/pAiSRkZGfJ6vTp69GiD31NTUyOv1xuwAQCAyNfkoFJfX6958+bpscceU9++fSVJHo9HsbGxio+PD2iblJQkj8fjb/N/Q8qN8zfONSQ/P18Oh8O/paSkNLVsAADQjDQ5qLjdbn377bfasGFDKOtpUF5enqqqqvzbmTNnwv6dAADAek1aR2XOnDnaunWrdu/erS5duviPO51O1dbWqrKyMmBUpaKiQk6n099m//79Ade78VTQjTY3s9lsstlsTSkVAAA0Y0EFFZ/Pp7lz52rjxo3atWuXunfvHnB+4MCBat26tQoLC5WdnS1JKi0tVVlZmVwulyTJ5XLpzTff1Pnz55WYmChJ2rlzp+x2u/r06ROKPgFo5pjYDOCGoIKK2+1WQUGBNm/erLi4OP+cEofDobZt28rhcGj69OnKzc1VQkKC7Ha75s6dK5fLpaFDh0qSRo8erT59+mjKlClatmyZPB6PFi1aJLfbzagJwoYnRgCgeQoqqKxevVqSNHz48IDja9eu1XPPPSdJWr58uaKjo5Wdna2amhplZGTonXfe8beNiYnR1q1bNXv2bLlcLrVv3145OTl644037qwnAAAg4gR96+fntGnTRqtWrdKqVatu26Zr16767LPPgvlqAADQAvGuHwAAYCyCCgAAMBZBBQAAGIugAgAAjEVQAQAAxiKoAAAAYxFUAACAsQgqAADAWAQVAABgrCa9PRkwGS+0A4DIwYgKAAAwFkEFAAAYi6ACAACMRVABAADGYjItWiQm3AJA88CICgAAMBZBBQAAGIugAgAAjEVQAQAAxmIyLZo9JsYCQOQiqABolhoKqKeXZllQCYBw4tYPAAAwFkEFAAAYi6ACAACMRVABAADGIqgAAABjEVQAAICxCCoAAMBYBBUAAGAsggoAADAWQQUAABiLoAIAAIxFUAEAAMYiqAAAAGMRVAAAgLEIKgAAwFgEFQAAYCyCCgAAMBZBBQAAGIugAgAAjNXK6gIAIFS6vfxpwP7ppVkWVQIgVBhRAQAAxiKoAAAAYxFUAACAsYIOKrt379a4ceOUnJysqKgobdq0KeD8c889p6ioqIBtzJgxAW0uXbqkyZMny263Kz4+XtOnT9eVK1fuqCMAACDyBD2Ztrq6WgMGDNDzzz+vCRMmNNhmzJgxWrt2rX/fZrMFnJ88ebLOnTunnTt36tq1a5o2bZpmzpypgoKCYMtBhGNyJAC0bEEHlczMTGVmZv5kG5vNJqfT2eC57777Ttu2bdOBAwc0aNAgSdLKlSs1duxYvfXWW0pOTg62JAAAEKHCMkdl165dSkxM1IMPPqjZs2fr4sWL/nPFxcWKj4/3hxRJSk9PV3R0tPbt29fg9WpqauT1egM2AAAQ+UIeVMaMGaMPPvhAhYWF+vOf/6yioiJlZmaqrq5OkuTxeJSYmBjwmVatWikhIUEej6fBa+bn58vhcPi3lJSUUJcNAAAMFPIF3yZNmuT/c79+/dS/f3/17NlTu3bt0qhRo5p0zby8POXm5vr3vV4vYaWFunnOCgAgsoV9ZdoePXro3nvv1YkTJzRq1Cg5nU6dP38+oM3169d16dKl285rsdlst0zIBYCf01CwZUI20LyEfR2Vs2fP6uLFi+rcubMkyeVyqbKyUiUlJf42X375perr65WWlhbucgAAQDMS9IjKlStXdOLECf/+qVOndOjQISUkJCghIUGvv/66srOz5XQ6dfLkSb300ku67777lJGRIUnq3bu3xowZoxkzZmjNmjW6du2a5syZo0mTJvHEDwAACBD0iMrBgwf1yCOP6JFHHpEk5ebm6pFHHtHixYsVExOjw4cP68knn9QDDzyg6dOna+DAgfr3v/8dcOtm/fr16tWrl0aNGqWxY8dq2LBh+tvf/ha6XgEAgIgQ9IjK8OHD5fP5bnt++/btP3uNhIQEFncDAAA/i3f9AAAAYxFUAACAsQgqAADAWAQVAABgLIIKAAAwFkEFAAAYi6ACAACMFfZ3/QCASW5+/w/v/gHMxogKAAAwFkEFAAAYi6ACAACMRVABAADGIqgAAABjEVQAAICxCCoAAMBYBBUAAGAsggoAADAWQQUAABiLoAIAAIxFUAEAAMYiqAAAAGMRVAAAgLEIKgAAwFgEFQAAYCyCCgAAMBZBBQAAGIugAgAAjEVQAQAAxmpldQHADd1e/tTqEgAAhmFEBQAAGIugAgAAjEVQAQAAxmKOCu4K5p/AVA39v3l6aZYFlQBoCCMqAADAWAQVAABgLIIKAAAwFkEFAAAYi6ACAACMRVABAADGIqgAAABjEVQAAICxCCoAAMBYrEyLsGAlWkQSVq8FrBP0iMru3bs1btw4JScnKyoqSps2bQo47/P5tHjxYnXu3Flt27ZVenq6vv/++4A2ly5d0uTJk2W32xUfH6/p06frypUrd9QRAAAQeYIOKtXV1RowYIBWrVrV4Plly5bp7bff1po1a7Rv3z61b99eGRkZunr1qr/N5MmTdfToUe3cuVNbt27V7t27NXPmzKb3AgAARKSgb/1kZmYqMzOzwXM+n08rVqzQokWL9NRTT0mSPvjgAyUlJWnTpk2aNGmSvvvuO23btk0HDhzQoEGDJEkrV67U2LFj9dZbbyk5OfkOugMAACJJSCfTnjp1Sh6PR+np6f5jDodDaWlpKi4uliQVFxcrPj7eH1IkKT09XdHR0dq3b18oywEAAM1cSCfTejweSVJSUlLA8aSkJP85j8ejxMTEwCJatVJCQoK/zc1qampUU1Pj3/d6vaEsGwAAGKpZPJ6cn58vh8Ph31JSUqwuCQAA3AUhDSpOp1OSVFFREXC8oqLCf87pdOr8+fMB569fv65Lly7529wsLy9PVVVV/u3MmTOhLBsAABgqpEGle/fucjqdKiws9B/zer3at2+fXC6XJMnlcqmyslIlJSX+Nl9++aXq6+uVlpbW4HVtNpvsdnvABgAAIl/Qc1SuXLmiEydO+PdPnTqlQ4cOKSEhQampqZo3b57++Mc/6v7771f37t316quvKjk5WePHj5ck9e7dW2PGjNGMGTO0Zs0aXbt2TXPmzNGkSZN44gcAAAQIOqgcPHhQI0aM8O/n5uZKknJycrRu3Tq99NJLqq6u1syZM1VZWalhw4Zp27ZtatOmjf8z69ev15w5czRq1ChFR0crOztbb7/9dgi6AwAAIkmUz+fzWV1EsLxerxwOh6qqqrgNZCiW0EdzdvPy+CyhD4RGU35/N4unfgAAQMtEUAEAAMYiqAAAAGMRVAAAgLEIKgAAwFgEFQAAYCyCCgAAMFZI354MAC3FzWursK4KEB4EFQC4CQsWAubg1g8AADAWQQUAABiLoAIAAIxFUAEAAMYiqAAAAGMRVAAAgLEIKgAAwFgEFQAAYCyCCgAAMBYr0yJoLB0OALhbGFEBAADGYkQFARgtAQCYhKCCO8YL3AAA4cKtHwAAYCyCCgAAMBZBBQAAGIugAgAAjEVQAQAAxiKoAAAAY/F4Mn4Sjx4DAKzEiAoAADAWQQUAABiLWz8AEAIN3Sa9+RUUjWkDIBAjKgAAwFgEFQAAYCyCCgAAMBZBBQAAGIvJtAAQJqxDBNw5RlQAAICxCCoAAMBYBBUAAGAsggoAADAWQQUAABiLoAIAAIzF48kRineKAAAiQchHVF577TVFRUUFbL169fKfv3r1qtxutzp27KgOHTooOztbFRUVoS4DAABEgLDc+nnooYd07tw5//b111/7z82fP19btmzRxx9/rKKiIpWXl2vChAnhKAMAADRzYbn106pVKzmdzluOV1VV6b333lNBQYFGjhwpSVq7dq169+6tvXv3aujQoeEoB7fBqpkAANOFZUTl+++/V3Jysnr06KHJkyerrKxMklRSUqJr164pPT3d37ZXr15KTU1VcXHxba9XU1Mjr9cbsAEAgMgX8qCSlpamdevWadu2bVq9erVOnTqlxx9/XJcvX5bH41FsbKzi4+MDPpOUlCSPx3Pba+bn58vhcPi3lJSUUJcNAAAMFPJbP5mZmf4/9+/fX2lpaeratas++ugjtW3btknXzMvLU25urn/f6/USVgAAaAHCvo5KfHy8HnjgAZ04cUJOp1O1tbWqrKwMaFNRUdHgnJYbbDab7HZ7wAYAACJf2NdRuXLlik6ePKkpU6Zo4MCBat26tQoLC5WdnS1JKi0tVVlZmVwuV7hLAQDjsQYSECjkQeXFF1/UuHHj1LVrV5WXl2vJkiWKiYnRs88+K4fDoenTpys3N1cJCQmy2+2aO3euXC4XT/wAaJF4+g74aSEPKmfPntWzzz6rixcvqlOnTho2bJj27t2rTp06SZKWL1+u6OhoZWdnq6amRhkZGXrnnXdCXQYAAIgAIQ8qGzZs+Mnzbdq00apVq7Rq1apQfzUAAIgwvJQQAAAYi6ACAACMRVABAADGIqgAAABjEVQAAICxCCoAAMBYBBUAAGAsggoAADAWQQUAABiLoAIAAIxFUAEAAMYiqAAAAGOF/KWEMBevkwcANDeMqAAAAGMRVAAAgLEIKgAAwFjMUYkQzD8BAEQiRlQAAICxGFEBAMPdPGJ6emmWRZUAdx8jKgAAwFgEFQAAYCxu/TQDDPsCAFoqRlQAAICxCCoAAMBY3PpphlgzBWjZGvoZwC1hRCpGVAAAgLEIKgAAwFjc+gGACMTtIUQKRlQAAICxGFExDBNlATRFqH52sG4TTENQAYAWIlQhhDCDu4mgchdxzxgAgOAwRwUAABiLoAIAAIzFrR8AaKEaMwG3qW24rY1QYUQFAAAYi6ACAACMxa0fAIARuIWEhjCiAgAAjEVQAQAAxuLWj8VYMh8AgNsjqIQIgQMAbo+fkWgqggoAIOQIJggVgkoT8Y8QAIDws3Qy7apVq9StWze1adNGaWlp2r9/v5XlAAAAw1g2ovLhhx8qNzdXa9asUVpamlasWKGMjAyVlpYqMTHRqrIkMVoCAGgc1n4JP8uCyl/+8hfNmDFD06ZNkyStWbNGn376qf7xj3/o5ZdftqosAEAz19TwcPPnQhU4CDN3xpKgUltbq5KSEuXl5fmPRUdHKz09XcXFxbe0r6mpUU1NjX+/qqpKkuT1esNSX33Nf8NyXQBAcBrzc77vku0/2yZ1/sdBf3dTPtPYzzXl91dD/fz29Yygr3O3r/1/3ei3z+dr9GcsCSo//vij6urqlJSUFHA8KSlJx48fv6V9fn6+Xn/99VuOp6SkhK1GAID1HCusriA8QtWvcP73Cee1L1++LIfD0ai2zeKpn7y8POXm5vr36+vrdenSJXXs2FFRUVF3fH2v16uUlBSdOXNGdrv9jq9nMvoauVpSf+lrZGpJfZVaVn9v9LWsrExRUVFKTk5u9GctCSr33nuvYmJiVFFREXC8oqJCTqfzlvY2m002my3gWHx8fMjrstvtEf8/yw30NXK1pP7S18jUkvoqtaz+OhyOoPtqyePJsbGxGjhwoAoLC/3H6uvrVVhYKJfLZUVJAADAQJbd+snNzVVOTo4GDRqkIUOGaMWKFaqurvY/BQQAAGBZUJk4caIuXLigxYsXy+Px6OGHH9a2bdtumWB7N9hsNi1ZsuSW20uRiL5GrpbUX/oamVpSX6WW1d876WuUL5hnhAAAAO4iS5fQBwAA+CkEFQAAYCyCCgAAMBZBBQAAGIugcpMnn3xSqampatOmjTp37qwpU6aovLzc6rJC7vTp05o+fbq6d++utm3bqmfPnlqyZIlqa2utLi0s3nzzTT366KNq165dWBYLtNqqVavUrVs3tWnTRmlpadq/f7/VJYXF7t27NW7cOCUnJysqKkqbNm2yuqSwyc/P1+DBgxUXF6fExESNHz9epaWlVpcVFqtXr1b//v39C5+5XC59/vnnVpd1VyxdulRRUVGaN2+e1aWExWuvvaaoqKiArVevXkFdg6BykxEjRuijjz5SaWmp/vnPf+rkyZN6+umnrS4r5I4fP676+nq9++67Onr0qJYvX641a9bolVdesbq0sKitrdUzzzyj2bNnW11KyH344YfKzc3VkiVL9M0332jAgAHKyMjQ+fPnrS4t5KqrqzVgwACtWrXK6lLCrqioSG63W3v37tXOnTt17do1jR49WtXV1VaXFnJdunTR0qVLVVJSooMHD2rkyJF66qmndPToUatLC6sDBw7o3XffVf/+/a0uJaweeughnTt3zr99/fXXwV3Ah5+0efNmX1RUlK+2ttbqUsJu2bJlvu7du1tdRlitXbvW53A4rC4jpIYMGeJzu93+/bq6Ol9ycrIvPz/fwqrCT5Jv48aNVpdx15w/f94nyVdUVGR1KXfFPffc4/v73/9udRlhc/nyZd/999/v27lzp+9//ud/fC+88ILVJYXFkiVLfAMGDLijazCi8hMuXbqk9evX69FHH1Xr1q2tLifsqqqqlJCQYHUZCEJtba1KSkqUnp7uPxYdHa309HQVFxdbWBlCraqqSpIi/t9oXV2dNmzYoOrq6oh+pYrb7VZWVlbAv91I9f333ys5OVk9evTQ5MmTVVZWFtTnCSoNWLhwodq3b6+OHTuqrKxMmzdvtrqksDtx4oRWrlyp3/3ud1aXgiD8+OOPqquru2VF56SkJHk8HouqQqjV19dr3rx5euyxx9S3b1+rywmLI0eOqEOHDrLZbJo1a5Y2btyoPn36WF1WWGzYsEHffPON8vPzrS4l7NLS0rRu3Tpt27ZNq1ev1qlTp/T444/r8uXLjb5GiwgqL7/88i2TeW7ejh8/7m+/YMEC/ec//9GOHTsUExOjqVOnytdMFvANtq+S9MMPP2jMmDF65plnNGPGDIsqD15T+go0R263W99++602bNhgdSlh8+CDD+rQoUPat2+fZs+erZycHB07dszqskLuzJkzeuGFF7R+/Xq1adPG6nLCLjMzU88884z69++vjIwMffbZZ6qsrNRHH33U6Gu0iCX0L1y4oIsXL/5kmx49eig2NvaW42fPnlVKSor27NnTLIYhg+1reXm5hg8frqFDh2rdunWKjm4+2bUpf6/r1q3TvHnzVFlZGebq7o7a2lq1a9dOn3zyicaPH+8/npOTo8rKyogeDYyKitLGjRsD+h2J5syZo82bN2v37t3q3r271eXcNenp6erZs6feffddq0sJqU2bNunXv/61YmJi/Mfq6uoUFRWl6Oho1dTUBJyLRIMHD1Z6enqjR5Qseynh3dSpUyd16tSpSZ+tr6+XJNXU1ISypLAJpq8//PCDRowYoYEDB2rt2rXNKqRId/b3GiliY2M1cOBAFRYW+n9h19fXq7CwUHPmzLG2ONwRn8+nuXPnauPGjdq1a1eLCinS//5/3Fx+7gZj1KhROnLkSMCxadOmqVevXlq4cGHEh5QrV67o5MmTmjJlSqM/0yKCSmPt27dPBw4c0LBhw3TPPffo5MmTevXVV9WzZ89mMZoSjB9++EHDhw9X165d9dZbb+nChQv+c06n08LKwqOsrEyXLl1SWVmZ6urqdOjQIUnSfffdpw4dOlhb3B3Kzc1VTk6OBg0apCFDhmjFihWqrq7WtGnTrC4t5K5cuaITJ07490+dOqVDhw4pISFBqampFlYWem63WwUFBdq8ebPi4uL8c44cDofatm1rcXWhlZeXp8zMTKWmpury5csqKCjQrl27tH37dqtLC7m4uLhb5hndmBMZifOPXnzxRY0bN05du3ZVeXm5lixZopiYGD377LONv8idP3wUOQ4fPuwbMWKELyEhwWez2XzdunXzzZo1y3f27FmrSwu5tWvX+iQ1uEWinJycBvv61VdfWV1aSKxcudKXmprqi42N9Q0ZMsS3d+9eq0sKi6+++qrBv8ecnByrSwu52/37XLt2rdWlhdzzzz/v69q1qy82NtbXqVMn36hRo3w7duywuqy7JpIfT544caKvc+fOvtjYWN8vfvEL38SJE30nTpwI6hotYo4KAABonprXpAQAANCiEFQAAICxCCoAAMBYBBUAAGAsggoAADAWQQUAABiLoAIAAIxFUAEAAMYiqAAAAGMRVAAAgLEIKgAAwFgEFQAAYKz/B/hmuYBGFG2BAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "crest_factor mean: 1.6 (0.5)\n",
      "crest_factor min: -41.9 max: 6.3\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "stereo_width mean: 0.1 (0.1)\n",
      "stereo_width min: -1.8 max: 7.6\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "spectral_flatness mean: 0.2 (0.1)\n",
      "spectral_flatness min: -2.5 max: 5.9\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAigAAAGdCAYAAAA44ojeAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy80BEi2AAAACXBIWXMAAA9hAAAPYQGoP6dpAAAkG0lEQVR4nO3df3BU1f3/8VcSyEIgu2mAZJMhIOIPiBDAgGHFMigxASKVMVpRhOgwUJkNLcQixKEg6IcwaAuKSLRjxc6QYrWiNQjIjxJU4q9Yhl+aGtQBgU2olF2Jw/Ij+/2jX3a6ECAbEvbs5vmYuTO59569933Zgbw499xzo3w+n08AAAAGiQ51AQAAAOcjoAAAAOMQUAAAgHEIKAAAwDgEFAAAYBwCCgAAMA4BBQAAGIeAAgAAjNMu1AU0R0NDgw4fPqz4+HhFRUWFuhwAANAEPp9PP/74o1JTUxUdfek+krAMKIcPH1ZaWlqoywAAAM1w8OBBde/e/ZJtwjKgxMfHS/rvBVqt1hBXAwAAmsLj8SgtLc3/e/xSwjKgnLutY7VaCSgAAISZpgzPYJAsAAAwDgEFAAAYh4ACAACMQ0ABAADGIaAAAADjEFAAAIBxCCgAAMA4BBQAAGAcAgoAADAOAQUAABiHgAIAAIxDQAEAAMYhoAAAAOMQUAAAgHHahboAIBSumbPugm3fLc4LQSUAgMYQUIArQNABgNZBQAFa2PmhhcACAMFjDAoAADAOAQUAABiHgAIAAIxDQAEAAMYhoAAAAOMEFVBWrlypjIwMWa1WWa1WORwOrV+/3r9/xIgRioqKClgeffTRgGMcOHBAeXl5iouLU1JSkmbNmqUzZ860zNUAAICIENRjxt27d9fixYt1/fXXy+fz6bXXXtPdd9+tf/7zn7rpppskSVOmTNHChQv9n4mLi/P/fPbsWeXl5clut2vHjh06cuSIJk2apPbt22vRokUtdEkAACDcBRVQxo4dG7D+f//3f1q5cqU+/vhjf0CJi4uT3W5v9PPvv/++9u3bp82bNys5OVkDBw7UU089pdmzZ+vJJ59UbGxsMy8DAABEkmaPQTl79qzWrFmj+vp6ORwO//bVq1era9eu6tevn4qLi/XTTz/591VWVqp///5KTk72b8vNzZXH49HevXsvei6v1yuPxxOwAACAyBX0TLK7d++Ww+HQyZMn1blzZ61du1bp6emSpAcffFA9e/ZUamqqdu3apdmzZ6u6ulpvvfWWJMnlcgWEE0n+dZfLddFzlpSUaMGCBcGWCgAAwlTQAeXGG2/Uzp075Xa79eabb6qgoEAVFRVKT0/X1KlT/e369++vlJQUjRw5Uvv371fv3r2bXWRxcbGKior86x6PR2lpac0+HtAUvGcHAEIn6Fs8sbGxuu6665SZmamSkhINGDBAzz33XKNts7KyJEk1NTWSJLvdrtra2oA259YvNm5FkiwWi//JoXMLAACIXFc8D0pDQ4O8Xm+j+3bu3ClJSklJkSQ5HA7t3r1bdXV1/jabNm2S1Wr13yYCAAAI6hZPcXGxRo8erR49eujHH39UWVmZtm3bpo0bN2r//v0qKyvTmDFj1KVLF+3atUszZ87U8OHDlZGRIUnKyclRenq6Jk6cqCVLlsjlcmnu3LlyOp2yWCytcoEAACD8BBVQ6urqNGnSJB05ckQ2m00ZGRnauHGj7rzzTh08eFCbN2/WsmXLVF9fr7S0NOXn52vu3Ln+z8fExKi8vFzTpk2Tw+FQp06dVFBQEDBvCgAAQFAB5ZVXXrnovrS0NFVUVFz2GD179tR7770XzGkBAEAbw7t4AACAcQgoAADAOAQUAABgHAIKAAAwDgEFAAAYh4ACAACMQ0ABAADGIaAAAADjBP02YyBSNfb2YgBAaNCDAgAAjENAAQAAxiGgAAAA4xBQAACAcQgoAADAODzFA7Syxp4O+m5xXggqAYDwQUBBm8AjxAAQXggoiDiEEQAIf4xBAQAAxiGgAAAA4xBQAACAcQgoAADAOAySBYLAAFwAuDroQQEAAMYhoAAAAOMQUAAAgHEIKAAAwDgEFAAAYBwCCgAAMA4BBQAAGIeAAgAAjENAAQAAxiGgAAAA4xBQAACAcQgoAADAOAQUAABgHAIKAAAwDgEFAAAYJ6iAsnLlSmVkZMhqtcpqtcrhcGj9+vX+/SdPnpTT6VSXLl3UuXNn5efnq7a2NuAYBw4cUF5enuLi4pSUlKRZs2bpzJkzLXM1AAAgIgQVULp3767FixerqqpKn3/+ue644w7dfffd2rt3ryRp5syZevfdd/XGG2+ooqJChw8f1j333OP//NmzZ5WXl6dTp05px44deu2117Rq1SrNmzevZa8KAACEtSifz+e7kgMkJibqmWee0b333qtu3bqprKxM9957ryTpq6++Ut++fVVZWamhQ4dq/fr1uuuuu3T48GElJydLkkpLSzV79mwdPXpUsbGxTTqnx+ORzWaT2+2W1Wq9kvIRga6Zsy7UJVzWd4vzQl0CAFx1wfz+bvYYlLNnz2rNmjWqr6+Xw+FQVVWVTp8+rezsbH+bPn36qEePHqqsrJQkVVZWqn///v5wIkm5ubnyeDz+XhgAAIB2wX5g9+7dcjgcOnnypDp37qy1a9cqPT1dO3fuVGxsrBISEgLaJycny+VySZJcLldAODm3/9y+i/F6vfJ6vf51j8cTbNkAACCMBB1QbrzxRu3cuVNut1tvvvmmCgoKVFFR0Rq1+ZWUlGjBggWteg4glBq7LcVtIABtWdC3eGJjY3XdddcpMzNTJSUlGjBggJ577jnZ7XadOnVKx48fD2hfW1sru90uSbLb7Rc81XNu/VybxhQXF8vtdvuXgwcPBls2AAAII1c8D0pDQ4O8Xq8yMzPVvn17bdmyxb+vurpaBw4ckMPhkCQ5HA7t3r1bdXV1/jabNm2S1WpVenr6Rc9hsVj8jzafWwAAQOQK6hZPcXGxRo8erR49eujHH39UWVmZtm3bpo0bN8pms2ny5MkqKipSYmKirFarpk+fLofDoaFDh0qScnJylJ6erokTJ2rJkiVyuVyaO3eunE6nLBZLq1wgAAAIP0EFlLq6Ok2aNElHjhyRzWZTRkaGNm7cqDvvvFOStHTpUkVHRys/P19er1e5ubl68cUX/Z+PiYlReXm5pk2bJofDoU6dOqmgoEALFy5s2asCIhDjVAC0JUEFlFdeeeWS+zt06KAVK1ZoxYoVF23Ts2dPvffee8GcFgAAtDG8iwcAABiHgAIAAIxDQAEAAMYJeqI2AFcuHN4XBAChRA8KAAAwDgEFAAAYh4ACAACMQ0ABAADGIaAAAADjEFAAAIBxCCgAAMA4zIOCsMIL8wCgbaAHBQAAGIeAAgAAjENAAQAAxmEMCozWlt9Z05avHQAIKAh7/CIHgMjDLR4AAGAcAgoAADAOAQUAABiHgAIAAIxDQAEAAMYhoAAAAOMQUAAAgHEIKAAAwDgEFAAAYBwCCgAAMA5T3QNh7Pxp/r9bnBeiSgCgZdGDAgAAjENAAQAAxiGgAAAA4xBQAACAcQgoAADAODzFA2Oc/0QKAKDtogcFAAAYh4ACAACMQ0ABAADGIaAAAADjBBVQSkpKNGTIEMXHxyspKUnjxo1TdXV1QJsRI0YoKioqYHn00UcD2hw4cEB5eXmKi4tTUlKSZs2apTNnzlz51QAAgIgQ1FM8FRUVcjqdGjJkiM6cOaMnnnhCOTk52rdvnzp16uRvN2XKFC1cuNC/HhcX5//57NmzysvLk91u144dO3TkyBFNmjRJ7du316JFi1rgkgAAQLgLKqBs2LAhYH3VqlVKSkpSVVWVhg8f7t8eFxcnu93e6DHef/997du3T5s3b1ZycrIGDhyop556SrNnz9aTTz6p2NjYZlwGAACIJFc0BsXtdkuSEhMTA7avXr1aXbt2Vb9+/VRcXKyffvrJv6+yslL9+/dXcnKyf1tubq48Ho/27t3b6Hm8Xq88Hk/AAgAAIlezJ2praGjQjBkzNGzYMPXr18+//cEHH1TPnj2VmpqqXbt2afbs2aqurtZbb70lSXK5XAHhRJJ/3eVyNXqukpISLViwoLmlAgCAMNPsgOJ0OrVnzx59+OGHAdunTp3q/7l///5KSUnRyJEjtX//fvXu3btZ5youLlZRUZF/3ePxKC0trXmFAwAA4zXrFk9hYaHKy8v1j3/8Q927d79k26ysLElSTU2NJMlut6u2tjagzbn1i41bsVgsslqtAQsAAIhcQQUUn8+nwsJCrV27Vlu3blWvXr0u+5mdO3dKklJSUiRJDodDu3fvVl1dnb/Npk2bZLValZ6eHkw5AAAgQgV1i8fpdKqsrEzvvPOO4uPj/WNGbDabOnbsqP3796usrExjxoxRly5dtGvXLs2cOVPDhw9XRkaGJCknJ0fp6emaOHGilixZIpfLpblz58rpdMpisbT8FQIAgLATVA/KypUr5Xa7NWLECKWkpPiX119/XZIUGxurzZs3KycnR3369NFjjz2m/Px8vfvuu/5jxMTEqLy8XDExMXI4HHrooYc0adKkgHlTAABA2xbl8/l8oS4iWB6PRzabTW63m/EoYeyaOetCXULE+W5xXqhLAICLCub3N+/iAQAAxiGgAAAA4xBQAACAcQgoAADAOM2eSRYIBgNiAQDBoAcFAAAYh4ACAACMQ0ABAADGIaAAAADjEFAAAIBxCCgAAMA4BBQAAGAcAgoAADAOAQUAABiHgAIAAIxDQAEAAMYhoAAAAOMQUAAAgHF4mzGuWGNvKv5ucV4IKgEARAp6UAAAgHEIKAAAwDgEFAAAYBwCCgAAMA4BBQAAGIeAAgAAjENAAQAAxiGgAAAA4xBQAACAcQgoAADAOEx1D0QQXjsAIFIQUIAIR2gBEI64xQMAAIxDQAEAAMYhoAAAAOMQUAAAgHEIKAAAwDgEFAAAYBwCCgAAME5QAaWkpERDhgxRfHy8kpKSNG7cOFVXVwe0OXnypJxOp7p06aLOnTsrPz9ftbW1AW0OHDigvLw8xcXFKSkpSbNmzdKZM2eu/GoAAEBECGqitoqKCjmdTg0ZMkRnzpzRE088oZycHO3bt0+dOnWSJM2cOVPr1q3TG2+8IZvNpsLCQt1zzz366KOPJElnz55VXl6e7Ha7duzYoSNHjmjSpElq3769Fi1a1PJXiJBobHIwAACaKsrn8/ma++GjR48qKSlJFRUVGj58uNxut7p166aysjLde++9kqSvvvpKffv2VWVlpYYOHar169frrrvu0uHDh5WcnCxJKi0t1ezZs3X06FHFxsZe9rwej0c2m01ut1tWq7W55aOFEEbCDzPJAgiFYH5/X9EYFLfbLUlKTEyUJFVVVen06dPKzs72t+nTp4969OihyspKSVJlZaX69+/vDyeSlJubK4/Ho7179zZ6Hq/XK4/HE7AAAIDI1eyA0tDQoBkzZmjYsGHq16+fJMnlcik2NlYJCQkBbZOTk+Vyufxt/jecnNt/bl9jSkpKZLPZ/EtaWlpzywYAAGGg2QHF6XRqz549WrNmTUvW06ji4mK53W7/cvDgwVY/JwAACJ1mvc24sLBQ5eXl2r59u7p37+7fbrfbderUKR0/fjygF6W2tlZ2u93f5tNPPw043rmnfM61OZ/FYpHFYmlOqQAAIAwF1YPi8/lUWFiotWvXauvWrerVq1fA/szMTLVv315btmzxb6uurtaBAwfkcDgkSQ6HQ7t371ZdXZ2/zaZNm2S1WpWenn4l1wIAACJEUD0oTqdTZWVleueddxQfH+8fM2Kz2dSxY0fZbDZNnjxZRUVFSkxMlNVq1fTp0+VwODR06FBJUk5OjtLT0zVx4kQtWbJELpdLc+fOldPppJcEAABICjKgrFy5UpI0YsSIgO2vvvqqHn74YUnS0qVLFR0drfz8fHm9XuXm5urFF1/0t42JiVF5ebmmTZsmh8OhTp06qaCgQAsXLryyKwEAABHjiuZBCRXmQTEL86CEH+ZBARAKV20eFAAAgNZAQAEAAMYhoAAAAOMQUAAAgHEIKAAAwDgEFAAAYBwCCgAAMA4BBQAAGIeAAgAAjENAAQAAxiGgAAAA4xBQAACAcQgoAADAOAQUAABgHAIKAAAwDgEFAAAYp12oC0D4uWbOulCXAACIcPSgAAAA4xBQAACAcQgoAADAOAQUAABgHAIKAAAwDgEFAAAYh4ACAACMQ0ABAADGIaAAAADjEFAAAIBxCCgAAMA4vIsHaIPOf5/Sd4vzQlQJADSOHhQAAGAcAgoAADAOAQUAABiHgAIAAIxDQAEAAMYhoAAAAOMQUAAAgHGYBwVAk5w/d4rE/CkAWk/QPSjbt2/X2LFjlZqaqqioKL399tsB+x9++GFFRUUFLKNGjQpoc+zYMU2YMEFWq1UJCQmaPHmyTpw4cUUXgpZxzZx1AQvahvO/d757AKEWdECpr6/XgAEDtGLFiou2GTVqlI4cOeJf/vKXvwTsnzBhgvbu3atNmzapvLxc27dv19SpU4OvHgAARKSgb/GMHj1ao0ePvmQbi8Uiu93e6L4vv/xSGzZs0GeffabBgwdLkpYvX64xY8bo2WefVWpqarAlAQCACNMqg2S3bdumpKQk3XjjjZo2bZp++OEH/77KykolJCT4w4kkZWdnKzo6Wp988kmjx/N6vfJ4PAELAACIXC0+SHbUqFG655571KtXL+3fv19PPPGERo8ercrKSsXExMjlcikpKSmwiHbtlJiYKJfL1egxS0pKtGDBgpYuFcAlMA4FQCi1eEAZP368/+f+/fsrIyNDvXv31rZt2zRy5MhmHbO4uFhFRUX+dY/Ho7S0tCuuFZfHLykAQCi0+jwo1157rbp27aqamhpJkt1uV11dXUCbM2fO6NixYxcdt2KxWGS1WgMWAAAQuVo9oHz//ff64YcflJKSIklyOBw6fvy4qqqq/G22bt2qhoYGZWVltXY5AAAgDAR9i+fEiRP+3hBJ+vbbb7Vz504lJiYqMTFRCxYsUH5+vux2u/bv36/HH39c1113nXJzcyVJffv21ahRozRlyhSVlpbq9OnTKiws1Pjx43mCBwAASGpGD8rnn3+uQYMGadCgQZKkoqIiDRo0SPPmzVNMTIx27dqlX/ziF7rhhhs0efJkZWZm6oMPPpDFYvEfY/Xq1erTp49GjhypMWPG6LbbbtPLL7/cclcFAADCWtA9KCNGjJDP57vo/o0bN172GImJiSorKwv21AAAoI3gZYEAAMA4vCywDeMRYgCAqehBAQAAxiGgAAAA4xBQAACAcQgoAADAOAQUAABgHAIKAAAwDgEFAAAYh4ACAACMQ0ABAADGIaAAAADjEFAAAIBxeBdPG8K7dwAA4YIeFAAAYBwCCgAAMA4BBQAAGIcxKACa7fxxTd8tzgtRJQAiDT0oAADAOAQUAABgHAIKAAAwDgEFAAAYh4ACAACMw1M8AFpMY7MV82QPgOagBwUAABiHgAIAAIxDQAEAAMYhoAAAAOMQUAAAgHEIKAAAwDg8ZhwheGkbACCS0IMCAACMQ0ABAADG4RZPhGpsRk8AAMIFPSgAAMA4BBQAAGAcAgoAADAOAQUAABgn6ICyfft2jR07VqmpqYqKitLbb78dsN/n82nevHlKSUlRx44dlZ2dra+//jqgzbFjxzRhwgRZrVYlJCRo8uTJOnHixBVdCAAAiBxBB5T6+noNGDBAK1asaHT/kiVL9Pzzz6u0tFSffPKJOnXqpNzcXJ08edLfZsKECdq7d682bdqk8vJybd++XVOnTm3+VQAAgIgS9GPGo0eP1ujRoxvd5/P5tGzZMs2dO1d33323JOnPf/6zkpOT9fbbb2v8+PH68ssvtWHDBn322WcaPHiwJGn58uUaM2aMnn32WaWmpl7B5QAAgEjQomNQvv32W7lcLmVnZ/u32Ww2ZWVlqbKyUpJUWVmphIQEfziRpOzsbEVHR+uTTz5p9Lher1cejydgAQAAkatFA4rL5ZIkJScnB2xPTk7273O5XEpKSgrY365dOyUmJvrbnK+kpEQ2m82/pKWltWTZAADAMGHxFE9xcbHcbrd/OXjwYKhLAgAArahFp7q32+2SpNraWqWkpPi319bWauDAgf42dXV1AZ87c+aMjh075v/8+SwWiywWS0uWCuAq4U3bAJqjRXtQevXqJbvdri1btvi3eTweffLJJ3I4HJIkh8Oh48ePq6qqyt9m69atamhoUFZWVkuWAwAAwlTQPSgnTpxQTU2Nf/3bb7/Vzp07lZiYqB49emjGjBl6+umndf3116tXr1763e9+p9TUVI0bN06S1LdvX40aNUpTpkxRaWmpTp8+rcLCQo0fP54neAAAgKRmBJTPP/9ct99+u3+9qKhIklRQUKBVq1bp8ccfV319vaZOnarjx4/rtttu04YNG9ShQwf/Z1avXq3CwkKNHDlS0dHRys/P1/PPP98ClwMAACJBlM/n84W6iGB5PB7ZbDa53W5ZrdZQl2OE8+/zA6ZiDArQdgXz+7tFB8midTDIEADQ1oTFY8YAAKBtIaAAAADjEFAAAIBxCCgAAMA4BBQAAGAcnuIBcFU19kg8T6YBOB89KAAAwDgEFAAAYBwCCgAAMA5jUMIQ09oDACIdPSgAAMA4BBQAAGAcAgoAADAOAQUAABiHgAIAAIxDQAEAAMYhoAAAAOMQUAAAgHEIKAAAwDjMJAsg5M6fHZm3GwOgBwUAABiHgAIAAIxDQAEAAMYhoAAAAOMQUAAAgHEIKAAAwDgEFAAAYBzmQTHM+fNBAADQFtGDAgAAjEMPCoCw1FhvIzPQApGDgAIgLHD7E2hbuMUDAACMQ0ABAADGIaAAAADjEFAAAIBxCCgAAMA4LR5QnnzySUVFRQUsffr08e8/efKknE6nunTpos6dOys/P1+1tbUtXYYRrpmzLmABAABN0yo9KDfddJOOHDniXz788EP/vpkzZ+rdd9/VG2+8oYqKCh0+fFj33HNPa5QBAADCVKvMg9KuXTvZ7fYLtrvdbr3yyisqKyvTHXfcIUl69dVX1bdvX3388ccaOnRoa5QDIMzQ4wigVQLK119/rdTUVHXo0EEOh0MlJSXq0aOHqqqqdPr0aWVnZ/vb9unTRz169FBlZWXEBxRmvgQAoGlaPKBkZWVp1apVuvHGG3XkyBEtWLBAP//5z7Vnzx65XC7FxsYqISEh4DPJyclyuVwXPabX65XX6/Wvezyeli4bAAAYpMUDyujRo/0/Z2RkKCsrSz179tRf//pXdezYsVnHLCkp0YIFC1qqRAAAYLhWfxdPQkKCbrjhBtXU1OjOO+/UqVOndPz48YBelNra2kbHrJxTXFysoqIi/7rH41FaWlprln3VcK8dAIALtfo8KCdOnND+/fuVkpKizMxMtW/fXlu2bPHvr66u1oEDB+RwOC56DIvFIqvVGrAAAIDI1eI9KL/97W81duxY9ezZU4cPH9b8+fMVExOjBx54QDabTZMnT1ZRUZESExNltVo1ffp0ORyOiB8gCwAAmq7FA8r333+vBx54QD/88IO6deum2267TR9//LG6desmSVq6dKmio6OVn58vr9er3Nxcvfjiiy1dBgAACGNRPp/PF+oiguXxeGSz2eR2u4253cNYEsA8PMYPmCWY39+8iwcAABiHgAIAAIxDQAEAAMYhoAAAAOMQUAAAgHEIKAAAwDgEFAAAYBwCCgAAMA4BBQAAGIeAAgAAjENAAQAAxiGgAAAA47T424wBwBSNvcSTFwgC4YEeFAAAYBwCCgAAMA63eAC0adwGAsxEQGmmxv5RAwAALYNbPAAAwDj0oABoU+j9BMIDPSgAAMA49KAAwGUwkBa4+uhBAQAAxony+Xy+UBcRLI/HI5vNJrfbLavV2urn4541gMuhRwW4vGB+f9ODAgAAjENAAQAAxiGgAAAA4xBQAACAcQgoAADAOMyDAgAGYc4V4L8IKADQSs4PGwQNoOkIKADQApgvCWhZjEEBAADGIaAAAADjcIunEXTVAgAQWgQUAAihlvoPEQNyEWm4xQMAAIxDDwoAXCWt1VsCRCICCgAYjkCCtiikAWXFihV65pln5HK5NGDAAC1fvly33HJLKEsCAPwPZrZFqIQsoLz++usqKipSaWmpsrKytGzZMuXm5qq6ulpJSUmhKgsA2gzCB0wWsoDyhz/8QVOmTNEjjzwiSSotLdW6dev0pz/9SXPmzAlVWQAQEVozfDTnllNTzk1gCh0T/+xDElBOnTqlqqoqFRcX+7dFR0crOztblZWVF7T3er3yer3+dbfbLUnyeDytUl+D96dWOS4AhFKPmW+0SJuWOveeBbkB643923v+v/P95m+87HEaa9MU5x+nKZpSTzhoyp99Szh3TJ/Pd/nGvhA4dOiQT5Jvx44dAdtnzZrlu+WWWy5oP3/+fJ8kFhYWFhYWlghYDh48eNmsEBZP8RQXF6uoqMi/3tDQoGPHjqlLly6KiooKYWXN4/F4lJaWpoMHD8pqtYa6HFwG31f44TsLL3xf4ae535nP59OPP/6o1NTUy7YNSUDp2rWrYmJiVFtbG7C9trZWdrv9gvYWi0UWiyVgW0JCQmuWeFVYrVb+MoYRvq/ww3cWXvi+wk9zvjObzdakdiGZSTY2NlaZmZnasmWLf1tDQ4O2bNkih8MRipIAAIBBQnaLp6ioSAUFBRo8eLBuueUWLVu2TPX19f6negAAQNsVsoBy//336+jRo5o3b55cLpcGDhyoDRs2KDk5OVQlXTUWi0Xz58+/4LYVzMT3FX74zsIL31f4uRrfWZTP15RnfQAAAK4e3mYMAACMQ0ABAADGIaAAAADjEFAAAIBxCCgh9N1332ny5Mnq1auXOnbsqN69e2v+/Pk6depUqEvD/1ixYoWuueYadejQQVlZWfr0009DXRIaUVJSoiFDhig+Pl5JSUkaN26cqqurQ10Wmmjx4sWKiorSjBkzQl0KLuHQoUN66KGH1KVLF3Xs2FH9+/fX559/3irnIqCE0FdffaWGhga99NJL2rt3r5YuXarS0lI98cQToS4N/9/rr7+uoqIizZ8/X1988YUGDBig3Nxc1dXVhbo0nKeiokJOp1Mff/yxNm3apNOnTysnJ0f19fWhLg2X8dlnn+mll15SRkZGqEvBJfznP//RsGHD1L59e61fv1779u3T73//e/3sZz9rlfPxmLFhnnnmGa1cuVLffPNNqEuBpKysLA0ZMkQvvPCCpP/OeJyWlqbp06drzpw5Ia4Ol3L06FElJSWpoqJCw4cPD3U5uIgTJ07o5ptv1osvvqinn35aAwcO1LJly0JdFhoxZ84cffTRR/rggw+uyvnoQTGM2+1WYmJiqMuApFOnTqmqqkrZ2dn+bdHR0crOzlZlZWUIK0NTuN1uSeLvk+GcTqfy8vIC/p7BTH//+981ePBg3XfffUpKStKgQYP0xz/+sdXOR0AxSE1NjZYvX65f/epXoS4Fkv7973/r7NmzF8xunJycLJfLFaKq0BQNDQ2aMWOGhg0bpn79+oW6HFzEmjVr9MUXX6ikpCTUpaAJvvnmG61cuVLXX3+9Nm7cqGnTpunXv/61XnvttVY5HwGlFcyZM0dRUVGXXL766quAzxw6dEijRo3SfffdpylTpoSociAyOJ1O7dmzR2vWrAl1KbiIgwcP6je/+Y1Wr16tDh06hLocNEFDQ4NuvvlmLVq0SIMGDdLUqVM1ZcoUlZaWtsr5QvYunkj22GOP6eGHH75km2uvvdb/8+HDh3X77bfr1ltv1csvv9zK1aGpunbtqpiYGNXW1gZsr62tld1uD1FVuJzCwkKVl5dr+/bt6t69e6jLwUVUVVWprq5ON998s3/b2bNntX37dr3wwgvyer2KiYkJYYU4X0pKitLT0wO29e3bV3/7299a5XwElFbQrVs3devWrUltDx06pNtvv12ZmZl69dVXFR1Np5YpYmNjlZmZqS1btmjcuHGS/vs/iC1btqiwsDC0xeECPp9P06dP19q1a7Vt2zb16tUr1CXhEkaOHKndu3cHbHvkkUfUp08fzZ49m3BioGHDhl3w6P6//vUv9ezZs1XOR0AJoUOHDmnEiBHq2bOnnn32WR09etS/j/+hm6GoqEgFBQUaPHiwbrnlFi1btkz19fV65JFHQl0azuN0OlVWVqZ33nlH8fHx/nFCNptNHTt2DHF1OF98fPwF44M6deqkLl26MG7IUDNnztStt96qRYsW6Ze//KU+/fRTvfzyy63W809ACaFNmzappqZGNTU1F3RF8/S3Ge6//34dPXpU8+bNk8vl0sCBA7Vhw4YLBs4i9FauXClJGjFiRMD2V1999bK3XAFc3pAhQ7R27VoVFxdr4cKF6tWrl5YtW6YJEya0yvmYBwUAABiHAQ8AAMA4BBQAAGAcAgoAADAOAQUAABiHgAIAAIxDQAEAAMYhoAAAAOMQUAAAgHEIKAAAwDgEFAAAYBwCCgAAMA4BBQAAGOf/AW5lWd4Ei1zNAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "silence_percentage mean: 0.3 (2.4)\n",
      "silence_percentage min: -0.1 max: 40.8\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "loudness mean: -inf (nan)\n",
      "loudness min: nan max: nan\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/christian/miniconda3/envs/suno_env/lib/python3.10/site-packages/matplotlib/axes/_axes.py:6973: RuntimeWarning: All-NaN slice encountered\n",
      "  xmin = min(xmin, np.nanmin(xi))\n",
      "/home/christian/miniconda3/envs/suno_env/lib/python3.10/site-packages/matplotlib/axes/_axes.py:6974: RuntimeWarning: All-NaN slice encountered\n",
      "  xmax = max(xmax, np.nanmax(xi))\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "autodetected range of [nan, nan] is not finite",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[25], line 20\u001b[0m\n\u001b[1;32m     17\u001b[0m \u001b[38;5;66;03m# min and max crest factor normalized\u001b[39;00m\n\u001b[1;32m     18\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mfeature\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m min: \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mdata[\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mfeature\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m_normalized\u001b[39m\u001b[38;5;124m'\u001b[39m]\u001b[38;5;241m.\u001b[39mmin()\u001b[38;5;132;01m:\u001b[39;00m\u001b[38;5;124m.1f\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m max: \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mdata[\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mfeature\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m_normalized\u001b[39m\u001b[38;5;124m'\u001b[39m]\u001b[38;5;241m.\u001b[39mmax()\u001b[38;5;132;01m:\u001b[39;00m\u001b[38;5;124m.1f\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m---> 20\u001b[0m \u001b[43mplt\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mhist\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdata\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43mf\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;132;43;01m{\u001b[39;49;00m\u001b[43mfeature\u001b[49m\u001b[38;5;132;43;01m}\u001b[39;49;00m\u001b[38;5;124;43m_normalized\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbins\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m100\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m     21\u001b[0m plt\u001b[38;5;241m.\u001b[39mshow()\n",
      "File \u001b[0;32m~/miniconda3/envs/suno_env/lib/python3.10/site-packages/matplotlib/pyplot.py:3354\u001b[0m, in \u001b[0;36mhist\u001b[0;34m(x, bins, range, density, weights, cumulative, bottom, histtype, align, orientation, rwidth, log, color, label, stacked, data, **kwargs)\u001b[0m\n\u001b[1;32m   3329\u001b[0m \u001b[38;5;129m@_copy_docstring_and_deprecators\u001b[39m(Axes\u001b[38;5;241m.\u001b[39mhist)\n\u001b[1;32m   3330\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mhist\u001b[39m(\n\u001b[1;32m   3331\u001b[0m     x: ArrayLike \u001b[38;5;241m|\u001b[39m Sequence[ArrayLike],\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   3352\u001b[0m     BarContainer \u001b[38;5;241m|\u001b[39m Polygon \u001b[38;5;241m|\u001b[39m \u001b[38;5;28mlist\u001b[39m[BarContainer \u001b[38;5;241m|\u001b[39m Polygon],\n\u001b[1;32m   3353\u001b[0m ]:\n\u001b[0;32m-> 3354\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mgca\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mhist\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   3355\u001b[0m \u001b[43m        \u001b[49m\u001b[43mx\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3356\u001b[0m \u001b[43m        \u001b[49m\u001b[43mbins\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mbins\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3357\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;28;43mrange\u001b[39;49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mrange\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3358\u001b[0m \u001b[43m        \u001b[49m\u001b[43mdensity\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdensity\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3359\u001b[0m \u001b[43m        \u001b[49m\u001b[43mweights\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mweights\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3360\u001b[0m \u001b[43m        \u001b[49m\u001b[43mcumulative\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcumulative\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3361\u001b[0m \u001b[43m        \u001b[49m\u001b[43mbottom\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mbottom\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3362\u001b[0m \u001b[43m        \u001b[49m\u001b[43mhisttype\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mhisttype\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3363\u001b[0m \u001b[43m        \u001b[49m\u001b[43malign\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43malign\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3364\u001b[0m \u001b[43m        \u001b[49m\u001b[43morientation\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43morientation\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3365\u001b[0m \u001b[43m        \u001b[49m\u001b[43mrwidth\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mrwidth\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3366\u001b[0m \u001b[43m        \u001b[49m\u001b[43mlog\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mlog\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3367\u001b[0m \u001b[43m        \u001b[49m\u001b[43mcolor\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcolor\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3368\u001b[0m \u001b[43m        \u001b[49m\u001b[43mlabel\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mlabel\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3369\u001b[0m \u001b[43m        \u001b[49m\u001b[43mstacked\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mstacked\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3370\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43m{\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mdata\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m:\u001b[49m\u001b[43m \u001b[49m\u001b[43mdata\u001b[49m\u001b[43m}\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mif\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mdata\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;129;43;01mis\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[38;5;129;43;01mnot\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[38;5;28;43;01melse\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43m{\u001b[49m\u001b[43m}\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3371\u001b[0m \u001b[43m        \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   3372\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/suno_env/lib/python3.10/site-packages/matplotlib/__init__.py:1473\u001b[0m, in \u001b[0;36m_preprocess_data.<locals>.inner\u001b[0;34m(ax, data, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1470\u001b[0m \u001b[38;5;129m@functools\u001b[39m\u001b[38;5;241m.\u001b[39mwraps(func)\n\u001b[1;32m   1471\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21minner\u001b[39m(ax, \u001b[38;5;241m*\u001b[39margs, data\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs):\n\u001b[1;32m   1472\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m data \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m-> 1473\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfunc\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1474\u001b[0m \u001b[43m            \u001b[49m\u001b[43max\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1475\u001b[0m \u001b[43m            \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;28;43mmap\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43msanitize_sequence\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1476\u001b[0m \u001b[43m            \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43m{\u001b[49m\u001b[43mk\u001b[49m\u001b[43m:\u001b[49m\u001b[43m \u001b[49m\u001b[43msanitize_sequence\u001b[49m\u001b[43m(\u001b[49m\u001b[43mv\u001b[49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mfor\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mk\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mv\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;129;43;01min\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mkwargs\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mitems\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m}\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1478\u001b[0m     bound \u001b[38;5;241m=\u001b[39m new_sig\u001b[38;5;241m.\u001b[39mbind(ax, \u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[1;32m   1479\u001b[0m     auto_label \u001b[38;5;241m=\u001b[39m (bound\u001b[38;5;241m.\u001b[39marguments\u001b[38;5;241m.\u001b[39mget(label_namer)\n\u001b[1;32m   1480\u001b[0m                   \u001b[38;5;129;01mor\u001b[39;00m bound\u001b[38;5;241m.\u001b[39mkwargs\u001b[38;5;241m.\u001b[39mget(label_namer))\n",
      "File \u001b[0;32m~/miniconda3/envs/suno_env/lib/python3.10/site-packages/matplotlib/axes/_axes.py:7001\u001b[0m, in \u001b[0;36mAxes.hist\u001b[0;34m(self, x, bins, range, density, weights, cumulative, bottom, histtype, align, orientation, rwidth, log, color, label, stacked, **kwargs)\u001b[0m\n\u001b[1;32m   6997\u001b[0m \u001b[38;5;66;03m# Loop through datasets\u001b[39;00m\n\u001b[1;32m   6998\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(nx):\n\u001b[1;32m   6999\u001b[0m     \u001b[38;5;66;03m# this will automatically overwrite bins,\u001b[39;00m\n\u001b[1;32m   7000\u001b[0m     \u001b[38;5;66;03m# so that each histogram uses the same bins\u001b[39;00m\n\u001b[0;32m-> 7001\u001b[0m     m, bins \u001b[38;5;241m=\u001b[39m \u001b[43mnp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mhistogram\u001b[49m\u001b[43m(\u001b[49m\u001b[43mx\u001b[49m\u001b[43m[\u001b[49m\u001b[43mi\u001b[49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbins\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mweights\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mw\u001b[49m\u001b[43m[\u001b[49m\u001b[43mi\u001b[49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mhist_kwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   7002\u001b[0m     tops\u001b[38;5;241m.\u001b[39mappend(m)\n\u001b[1;32m   7003\u001b[0m tops \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39marray(tops, \u001b[38;5;28mfloat\u001b[39m)  \u001b[38;5;66;03m# causes problems later if it's an int\u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda3/envs/suno_env/lib/python3.10/site-packages/numpy/lib/histograms.py:780\u001b[0m, in \u001b[0;36mhistogram\u001b[0;34m(a, bins, range, density, weights)\u001b[0m\n\u001b[1;32m    680\u001b[0m \u001b[38;5;250m\u001b[39m\u001b[38;5;124mr\u001b[39m\u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m    681\u001b[0m \u001b[38;5;124;03mCompute the histogram of a dataset.\u001b[39;00m\n\u001b[1;32m    682\u001b[0m \n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    776\u001b[0m \n\u001b[1;32m    777\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m    778\u001b[0m a, weights \u001b[38;5;241m=\u001b[39m _ravel_and_check_weights(a, weights)\n\u001b[0;32m--> 780\u001b[0m bin_edges, uniform_bins \u001b[38;5;241m=\u001b[39m \u001b[43m_get_bin_edges\u001b[49m\u001b[43m(\u001b[49m\u001b[43ma\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbins\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mrange\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mweights\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    782\u001b[0m \u001b[38;5;66;03m# Histogram is an integer or a float array depending on the weights.\u001b[39;00m\n\u001b[1;32m    783\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m weights \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n",
      "File \u001b[0;32m~/miniconda3/envs/suno_env/lib/python3.10/site-packages/numpy/lib/histograms.py:426\u001b[0m, in \u001b[0;36m_get_bin_edges\u001b[0;34m(a, bins, range, weights)\u001b[0m\n\u001b[1;32m    423\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m n_equal_bins \u001b[38;5;241m<\u001b[39m \u001b[38;5;241m1\u001b[39m:\n\u001b[1;32m    424\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m`bins` must be positive, when an integer\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m--> 426\u001b[0m     first_edge, last_edge \u001b[38;5;241m=\u001b[39m \u001b[43m_get_outer_edges\u001b[49m\u001b[43m(\u001b[49m\u001b[43ma\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mrange\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m    428\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m np\u001b[38;5;241m.\u001b[39mndim(bins) \u001b[38;5;241m==\u001b[39m \u001b[38;5;241m1\u001b[39m:\n\u001b[1;32m    429\u001b[0m     bin_edges \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39masarray(bins)\n",
      "File \u001b[0;32m~/miniconda3/envs/suno_env/lib/python3.10/site-packages/numpy/lib/histograms.py:323\u001b[0m, in \u001b[0;36m_get_outer_edges\u001b[0;34m(a, range)\u001b[0m\n\u001b[1;32m    321\u001b[0m     first_edge, last_edge \u001b[38;5;241m=\u001b[39m a\u001b[38;5;241m.\u001b[39mmin(), a\u001b[38;5;241m.\u001b[39mmax()\n\u001b[1;32m    322\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (np\u001b[38;5;241m.\u001b[39misfinite(first_edge) \u001b[38;5;129;01mand\u001b[39;00m np\u001b[38;5;241m.\u001b[39misfinite(last_edge)):\n\u001b[0;32m--> 323\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[1;32m    324\u001b[0m             \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mautodetected range of [\u001b[39m\u001b[38;5;132;01m{}\u001b[39;00m\u001b[38;5;124m, \u001b[39m\u001b[38;5;132;01m{}\u001b[39;00m\u001b[38;5;124m] is not finite\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;241m.\u001b[39mformat(first_edge, last_edge))\n\u001b[1;32m    326\u001b[0m \u001b[38;5;66;03m# expand empty range to avoid divide by zero\u001b[39;00m\n\u001b[1;32m    327\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m first_edge \u001b[38;5;241m==\u001b[39m last_edge:\n",
      "\u001b[0;31mValueError\u001b[0m: autodetected range of [nan, nan] is not finite"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# let's look at crest factor\n",
    "# compute mean and std to normalize the data\n",
    "# then lets print the filepath of audio\n",
    "# near the mean, one std, and two stds away\n",
    "\n",
    "for feature in ['bass', 'mid', 'high', 'crest_factor', 'stereo_width', 'spectral_flatness', 'silence_percentage', 'loudness']:\n",
    "    #plt.hist(data[feature], bins=100)\n",
    "    #plt.show()\n",
    "\n",
    "    mean = data[feature].mean()\n",
    "    std = data[feature].std()\n",
    "\n",
    "    print(f\"{feature} mean: {mean:.1f} ({std:.1f})\")\n",
    "\n",
    "    # add a column to the data for std normalized crest factor\n",
    "    data[f'{feature}_normalized'] = (data[feature] - mean) / std\n",
    "    # min and max crest factor normalized\n",
    "    print(f\"{feature} min: {data[f'{feature}_normalized'].min():.1f} max: {data[f'{feature}_normalized'].max():.1f}\")\n",
    "\n",
    "    plt.hist(data[f'{feature}_normalized'], bins=100)\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Bass mean: 167.1 (51.2)\n",
      "Mid mean: 117.6 (24.2)\n",
      "High mean: 46.2 (14.6)\n"
     ]
    }
   ],
   "source": [
    "# print the mean and std of the bass and high features\n",
    "print(f\"Bass mean: {data['bass'].mean():.1f} ({data['bass'].std():.1f})\")\n",
    "print(f\"Mid mean: {data['mid'].mean():.1f} ({data['mid'].std():.1f})\")\n",
    "print(f\"High mean: {data['high'].mean():.1f} ({data['high'].std():.1f})\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "417.1603088378906\n",
      "bass is 250.1 stds away from the mean\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/cbfd2e00-6f6d-43f8-8160-f3a3187d71b7.wav\n"
     ]
    }
   ],
   "source": [
    "# find the filepath with the highest bass\n",
    "feature = 'bass'\n",
    "print(data[feature].max())\n",
    "# measure how many standard deviations away from the mean \n",
    "print(f\"{feature} is {data[feature].max() - data[feature].mean():.1f} stds away from the mean\")\n",
    "filepath = data[data[feature] == data[feature].max()]['filepath'].values[0]\n",
    "print(filepath)\n",
    "\n",
    "# load the audio\n",
    "#x, sr = torchaudio.load(filepath)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAG1CAYAAAAfhDVuAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy80BEi2AAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOz9e3hk6V3fi37Wfa26V+mubkndM91zt2dsz3gwmBjjyTbYMZfAYTvxJt6EBydwHMLD3TsxBIK3Dz4cDhgI3jnPszHZB5JDEvBOHDDbwYYxvozn4rE9175M3yS1VJLqXrXu6z1/LFWppJbU6p5Wt1rzfp5HT5dab616a1W96/2u31URQggkEolEIpFIDinqrZ6ARCKRSCQSyX4ixY5EIpFIJJJDjRQ7EolEIpFIDjVS7EgkEolEIjnUSLEjkUgkEonkUCPFjkQikUgkkkONFDsSiUQikUgONVLsSCQSiUQiOdRIsSORSCQSieRQI8WORCKRSCSSQ80tFTuPP/4473nPe5ienkZRFD71qU/tOPaf/tN/iqIo/NZv/dam/6/Varzvfe+jUChQKpX40R/9UTqdzv5OXCKRSCQSyW3DLRU73W6XBx98kN/7vd/bddyf/dmf8ZWvfIXp6ekr/va+972P559/ns9+9rN8+tOf5vHHH+cDH/jAfk1ZIpFIJBLJbYZ+K1/8u7/7u/nu7/7uXccsLCzwz/7ZP+Mv//Ivefe7373pby+++CKf+cxnePLJJ3n44YcB+J3f+R3e9a538Ru/8RvbiqPtSJKExcVF8vk8iqJc35uRSCQSiURyUxFC0G63mZ6eRlV3tt/cUrFzNZIk4Yd/+If5uZ/7Oe6///4r/v7lL3+ZUqk0EDoAjz32GKqq8sQTT/D93//92x7X93183x/8vrCwwH333Xfj34BEIpFIJJJ959KlSxw9enTHvx9osfPrv/7r6LrOT/7kT27796WlJcbHxzf9n67rVCoVlpaWdjzuRz/6UX7lV37liv+/dOkShULh1U1asjvdLvQtbouLkM3e2vlIJIcdueYkh5hWq8XMzAz5fH7XcQdW7Dz99NP89m//Ns8888wNdy196EMf4qd/+qcHv/dPVqFQkGJnv9G0jceFgrzwSiT7jVxzktcAV9MJBzb1/Atf+ALVapXZ2Vl0XUfXdS5cuMDP/MzPcOzYMQAmJyepVqubnhdFEbVajcnJyR2PbVnWQNhIgSORSCQSyeHmwFp2fviHf5jHHnts0/+9853v5Id/+If5kR/5EQDe8pa30Gg0ePrpp3nTm94EwOc+9zmSJOHRRx+96XOWSCQSiURy8LilYqfT6XDmzJnB7+fOnePZZ5+lUqkwOzvLyMjIpvGGYTA5Ocndd98NwL333st3fdd38WM/9mN84hOfIAxDPvjBD/Le9753z5lYkpuMrsP737/xWCKR7C9yzUkkt1bsPPXUU7z97W8f/N6Po3n/+9/PJz/5yT0d44/+6I/44Ac/yDve8Q5UVeUHfuAH+PjHP74f05XcCCwL9vjZSiSSG4BccxIJihBC3OpJ3GparRbFYpFmsynjdyQSiUQiuU3Y6/4tbZqSm4sQ0OuljzMZkEUcJZL9Ra45ieTgZmNJDim9HuRy6U//AiyRSPYPueYkEil2JBKJRCKRHG6k2JFIJBKJRHKokWJHIpFIJBLJoUaKHYlEIpFIJIcaKXYkEolEIpEcaqTY2UeEEHhhjCxlJJFIJBLJrUPW2dknhBCcXu6w3PaYyNucnMjd8O7ttyWaBj/4gxuPJRLJ/iLXnEQixc5+4UcJy22PlhsBHrMjGWxDXmiwbfiP//FWz0Iiee0g15xEIsXOfmHpKhN5G0gtO5YuPYYSiUQikdwKpNjZJxRF4eREjtmRDJauSheWRCKRSCS3CGlu2EcURcE2NCl0hul20948ipI+lkgk+4tccxKJFDsSiUQikUgON1LsSCQSiUQiOdRIsSORSCQSieRQI8WORCKRSCSSQ40UOxKJRCKRSA41UuxIJBKJRCI51Mg6O5Kbi6bBu9618Vgikewvcs1JJFLsSG4ytg3/7b/d6llIJK8d5JqTSKQbSyKRSCQSyeFGih2JRCKRSCSHGil2JDeXbhey2fRHlq6XSPYfueYkEhmzI7kF9Hq3egYSyWsLueYkr3GkZUcikUgkEsmhRoodiUQikUgkhxopdiQSiUQikRxqpNiRSCQSiURyqJFiRyKRSCQSyaFGZmNJbi6qCm9728ZjiUSyv8g1J5FIsSO5yTgO/PVf3+pZSCSvHeSak0ikG0sikUgkEsnhRoodiUQikUgkhxopdiQ3l24XxsbSH1m6XiLZf+Sak0hkzI7kFrC6eqtnIJG8tpBrTvIaR1p2JBKJRCKRHGqk2JFIJBKJRHKokWJHIpFIJBLJoeaWip3HH3+c97znPUxPT6MoCp/61KcGfwvDkF/4hV/gda97Hdlslunpaf7RP/pHLC4ubjpGrVbjfe97H4VCgVKpxI/+6I/S6XRu8juRSCQSiURyULmlYqfb7fLggw/ye7/3e1f8rdfr8cwzz/DhD3+YZ555hj/90z/l5Zdf5nu+53s2jXvf+97H888/z2c/+1k+/elP8/jjj/OBD3zgZr0FyS1CCIEXxgghbvVUJBKJRHLAUcQB2S0UReHP/uzP+L7v+74dxzz55JO8+c1v5sKFC8zOzvLiiy9y33338eSTT/Lwww8D8JnPfIZ3vetdzM/PMz09vafXbrVaFItFms0mhULhRrwdyU64Lvydv5M+fvzxtLrrNSKE4PRyh+W2x0Te5uREDkVRbvBEJZJDwg1YcxLJQWWv+/dtFbPTbDZRFIVSqQTAl7/8ZUql0kDoADz22GOoqsoTTzxxi2Yp2RXHgSefTH+u86LrRwnLbY+WG7Hc9vCj5AZPUiI5RNyANSeR3O7cNnV2PM/jF37hF/gH/+AfDNTb0tIS4+Pjm8bpuk6lUmFpaWnHY/m+j+/7g99brdb+TFqyL1i6ykTeBlLLjqXfVppdIpFIJDeZ22KXCMOQH/qhH0IIwe///u+/6uN99KMfpVgsDn5mZmZuwCwlNwtFUTg5keORYxXpwpJIJBLJVTnwYqcvdC5cuMBnP/vZTT65yclJqtXqpvFRFFGr1ZicnNzxmB/60IdoNpuDn0uXLu3b/CVb6PXg2LH0p9e77sMoioJtaFLoSCRX4watOYnkduZAu7H6Quf06dN8/vOfZ2RkZNPf3/KWt9BoNHj66ad505veBMDnPvc5kiTh0Ucf3fG4lmVhWda+zl2yA0LAhQsbjyUSyf4i15xEcmvFTqfT4cyZM4Pfz507x7PPPkulUmFqaoof/MEf5JlnnuHTn/40cRwP4nAqlQqmaXLvvffyXd/1XfzYj/0Yn/jEJwjDkA9+8IO8973v3XMmlkQikUgkksPNLU09/+u//mve/va3X/H/73//+/lX/+pfcfz48W2f9/nPf57v+I7vANKigh/84Af5r//1v6KqKj/wAz/Axz/+cXK53J7nIVPPbyLdLvQ/m04HstlbOx+J5LAj15zkELPX/fuWWna+4zu+Y9eicHvRYZVKhT/+4z++kdM6FAgh8KMES1dvSFzLjT6eRCKRSCQ3iwMdsyO5Pm500T1ZxE8ikUgktzMHPhtLcu3c6KJ7soifRCKRSG5npGXnEHKji+7d0OMpCtx338ZjiUSyv8g1J5EcnN5Yt5LDGKAsY3YkEolEcti5LQKUJftHv+jeQT2eRCKRSCQ3CxmzI5FIJBKJ5FAjxY7k5tLrwf33pz+ydL1Esv/INSeRSDeW5CYjBLzwwsZjiUSyv8g1J5FIy45EIpFIJJLDjRQ7EolEIpFIDjVS7EgkEolEIjnUSLEjkUgkEonkUCPFzgFHCIEXxntqiiqRSCQSieRKZDbWAeZQNuBUFJib23gskUj2F7nmJBIpdg4yww04wWN2JHP7VzHOZOD8+Vs9C4nktYNccxKJFDsHmRvd0FMikUgkktciUuwcYBRF4eREjtmRzHU14JTNO28M8jxKJBLJ7Y00FRxw+g04r0fonF7u8OT5GqeXOwcnwNl14ZFH0h/XvdWzuSoH9jxKJHvlNltzEsl+IC07h5QDG++TJPDUUxuPDzg3+jxKK5HkpnObrTmJZD+QYueQIuN9bgw38jweyuw6iUQiuQ2QYueQ8mrjfSQpN/I8Hlhrm0QikRxypNi5zbgWN0g/3kfy6rhR51Fa2yQSieTWIMXObcRBcYMc1LiTgzqvPtLaJpFIJLcGKXZuIw6CG+SgCK7bZV5bkdY2iUQiuflIsXMbcRDcIDdEcI2OHsx5HWIOutVLss/sw5qTSG4npNi5jTgIbpBXLbiyWVhZOXjzOsTcLlYvyT6xT2tOIrmdkGLnNuNWu0FuteDayUJxq+d1kJFWL4lE8lpHih3JNXOrBNfVLBS3WggeVKTVSyKRvNaRV71DjhACL4xvWZuDK17fdeE7viP92aV0/XbzHrZQLLc9/EhWg90LfavXI8cq0oX1WmSPa04iOcxIy84h5lbHamz7+kkCf/M36YAdStfvNG9pobh+pNXrNcwe1pxEctiRYucQc6tjNbZ9/et93nozVBmXI5FIJJJrRYqdQ8yttoRs+/rBdT5vHWmhkEgkEsm1IsXOIeZWW0Ku9/Vv9bwlEolEcriQQQ+HnL4l5FYJhut9/Vs9793Y76DvV3v8Wx2ULpFIJAcNadmRSK6B/Q76frXHv9VB6RKJRHIQkZadQ4gQAjeI8ILoYN7dZzLpz23Ifqe/v9rjy/R8ybbcxmtOIrkRSMvOIUMIwamlNs9crAPwxrkyd03kD87dfTYL3e6rPsyt6vW030Hfr/b4tzooXXIAuUFrTiK5nZFi55DhRwnzDZeFhgcIxgs2cyPZQ5XBdCtdNfsdPL3T8fcq7mRwt0QikVyJFDuHDEtXOVpyqLY8AI6WnX2/u7/ZVpZbXT9ov9Pftx7/WsWdTM+XSCSSzdxSG/fjjz/Oe97zHqanp1EUhU996lOb/i6E4Jd+6ZeYmprCcRwee+wxTp8+vWlMrVbjfe97H4VCgVKpxI/+6I/S6XRu4rs4WCiKwl2Teb7vDUf4/jcc2XcXVn8jfvJ8jdPLnavHCHkevPvd6Y/nXddr9l01BUd/TbhqDnscjswe22duwJqTSG53buku0e12efDBB/m93/u9bf/+sY99jI9//ON84hOf4IknniCbzfLOd74Tb2jBvu997+P555/ns5/9LJ/+9Kd5/PHH+cAHPnCz3sKBRFEUHFPHNvV9t7RcbSO+YiOLY/jzP09/4njH4w4/b+sxXmu9ng6juOt/pkmSXJtYllw7e1xzEslhRhEH5OqiKAp/9md/xvd93/cB6cVwenqan/mZn+Fnf/ZnAWg2m0xMTPDJT36S9773vbz44ovcd999PPnkkzz88MMAfOYzn+Fd73oX8/PzTE9P7+m1W60WxWKRZrNJoVDYl/d3WNnNxbLt33o9yOXSJ3c6afDkLsccz1mgQLXt37JU6mtx0+2XS+9WBWTvB8Ofb8kxqPcC2l5MwdF55FhlVxfcYToPN41u96prTiK5Xdnr/n1gbxHPnTvH0tISjz322OD/isUijz76KF/+8pcB+PKXv0ypVBoIHYDHHnsMVVV54oknbvqcX4sMW1lOjGfxo2Rwd74X98vVupvPN1zm6y4tN2Kp5dJyw6ve/d9It8i1uOmu2aV3DRzkIovXyvDnW+8FVDLWnqxW+3l+JRLJ4ebABigvLS0BMDExsen/JyYmBn9bWlpifHx80991XadSqQzGbIfv+/i+P/i91WrdqGm/phi+y7Z09QorztV6Y21n+Vn/Q2rRgYFlZ7nl4QUJX59vMFlwdrTw3OhMrWsJhr7VgdO3C1u/FyfGswSxuKq1Rp5fiURyvRxYsbOffPSjH+VXfuVXbvU0bjv64sbUFPww5nS1S8MNmCw4zFScbTei3dKg083LHzxnpuJwqeam7qu8xcNz5cFmNlmw+fp8g7YXoygbxxdC4IcxArAN7YZviDvVrdnOnSJr3OyN7dLj7T2cKnl+JRLJ9XJgxc7k5CQAy8vLTE1NDf5/eXmZhx56aDCmWq1uel4URdRqtcHzt+NDH/oQP/3TPz34vdVqMTMzcwNnf/joW0yWWi5uELPUdLlQc5kq2ADMVpxtN6Ld0qC3bl4KDAkVmBvJDoREwTGYKNiEsct43sLS1bSA4nKbZy6sF1CcLXNyInfdG+J2Ama7jTlJEp5fbFHvBZusTIepxs1+x8ZcT3r8YTq/Eonk5nJgxc7x48eZnJzkr/7qrwbiptVq8cQTT/DjP/7jALzlLW+h0Wjw9NNP86Y3vQmAz33ucyRJwqOPPrrjsS3LwrKsfX8Ph4m+xWStE3Cp1qPrh7hBjBvGlDMm1lWsONuRbl7ZwXOA3YWK2PyvHyXM110WGi6gMF5wmRvNXteGuJv7a3hjFkLwwmKbL5xZxVDTvw9bjw5DjZuD3F/rMJxfiURy87mlYqfT6XDmzJnB7+fOnePZZ5+lUqkwOzvLT/3UT/Frv/ZrnDx5kuPHj/PhD3+Y6enpQcbWvffey3d913fxYz/2Y3ziE58gDEM++MEP8t73vnfPmViSvdG3wgghyJgaa5005ul1R0rcP10YbIZX3YiyWVgPLN3OerCTUPGjhGrHxwvTf+eibFpAsTxUQLHkbLhFrnFD3Kv7y48Saj0fQ1Vwg5iyYx46d4qMjTlkDK05ieS1yi0VO0899RRvf/vbB7/3XUvvf//7+eQnP8nP//zP0+12+cAHPkCj0eCtb30rn/nMZ7Bte/CcP/qjP+KDH/wg73jHO1BVlR/4gR/g4x//+E1/L/vNrU65HXYhmJpCECWDOJnrmc9O1oOdhMqG2HIpOQamlo69ayLPXCXzquYyfPyrub8sXWWiYDNf62HpKoa2fZD07ZweLWNjJBLJYePA1Nm5lRz0OjsH2a1wvXhhzJPna7TcaNf6KsPCoe9CqvX8TbEyW8XF8O/ADa+R4wYRXzq7hhvEFDPGprnv9bPa6bUOilA6KPOQSCSS3djr/n1gY3YkGxwqt4LnwQ//MBYw+f/8N8DO9VW2CofZikPdDTZlZG1NeT8xnuVMtTvI6EJAtbO3goR7dX/ZhsZMOTN4zeG57+Wz2kkQvVqhdCORsTGHiPU1B8D/8X/AkGVcInmtIMXOLWSvm9ahcivEMfyn/4QCnPiDP2DGcnZ8/1uFw8w2GV9+lLDUclnrBAghmChYg+cE60UMvTDhRorE3bKCtn5WpqbghfGmcTsJouH/F8JlomBRcIxNx0+SZFvr1k4cdAuS5CawvuYA+OQnb+lUJJJbhRQ7t4hrcU0d1pTbq1kPthN5s5W0pk8/PsdQoe1GXG66ZE2dnKUNnrPVsnMjReJOc98a29S3Mg1/xjuJ1+G4pO0KKAoheH6xtZ4Jlj5npuIMjglc4c57NRYkiUQiOSxIsXOLuFbX1GvRrTAQDhWHRAhOVzubemQJIXjhcpvztS4qCpahECabM7qAQebWzdrQ+5+VF8bbfsY7idf+/08UrEEBRdiw8PhhTLXlYSgKYZJQdkwu1npU2/62LrvtLGOKooAQh8ctKpFIJHtAip1bxKFyTW3DjXSTXKy5XKr3qLY88o4BpLVthBAstzwMVSESMJK1t009v9pGLoTACyL8KKHgGKjqjfksdvuMd7MMFRyDyYIDbFh4JvI2AkG17ZOzde6fLnJiPMtTFxs7uuyGX388Z20Io5yViiM4lN89iUQi2YoUO7eIw+qa6rdvuLDW23Ng8G70rRNuEAPgGBrjeQuRJFyo9Vhp++QsnQeOFLlvOn/NryOE4OWlFp95bomWF/HosQp/9/6JTYLneoXbXj7j3ao2D1t4wtgFIG8bOGZawNE2dnfZDb++SBK+9EoNN0zP48NzZeZGbq7FSyKRSG4VUuzcQg6ba6ofCzKwwtgbVpjrfZ8b1gk4OZ66tC7WXL70So1qyyNn62RMnbsm8tdlkfGjhPNrPV5Z7eBHgpeX23zLnSMUM+am93S98S27fcZXq9rct/AoijdoiNp3423nDoMrXXb9eJ5TS91B8cWT47lD00H9dkYGiUskNw8pdiTXfNHdafzACrNuPXBM7VW7SbZu6P1Kyv3XyJg6M+UM1h5cVdvN2dJVjo1kuGM0R8uLuHsiT97eWBb7mfZ/tWNvK2ZGrhQzV3PZ9c9Z3yo0W8ns+Dkf5A34IM/tWpFB4hLJzUWKndc413rRHR4/nreYrWwE3V5phclcaUHIZKDT2Xi8B4Y39D29xpb5DrvVxnMWsxUH1o+pKAp3TxaYq2S2jdnZSyr5tSKEwAtjEOKqsTPXGn+0HcPnrG8V2mleB3UDPshzux5uau2s61hzEslhQ4qd1zjXetHtj2+6IfO1HvN1l5lyZrD5XDUOSVHSXj3XybXEOm11q+UsnUtrXZ44JzA0lTfOlrlrMo3zcSwDZ5vesHtJJd/udXeyQAghOLXU5pmLaaf2N8yWeGSujLWPbqW9nrNXswHvt9XlUBXW5CYnKLzKNSeRHAak2HmN0t+cTE25potu/yIdRAltN+18vtzenFa935vQXl9jq1vN0FXiBJbbPspQl/ThVg/bbdhXSyUf5moWCD9KmG+4LDQ8QDBesDk2unnMfgiHq50zIQRiD5am7eZ2M6wut3P24m5B6IctQUEiOahIsfMaZGsV3hPj2T1fdIdr3wxnXO3ZveP78E/+Sfr4f/vfwNrGnHIN7CYMtrq8ZsoOF2s9vnaxAWx0Sd9rBtleNtyrWSAsXeVoaahTe9nZdJxXKxyuRyhtck3mrB0tTTvN7WZYXW5XcXC1IPSbYp26wWtOIrkdkWLnNcZ2VXivdXNSFAXb1Dk5kWOyaJOztD25dwCIIvjDP0wf/97vvaoL79WEwXYb5N2TBY6NZAdd0oGBq2u56eKYOkKIbc/JXjbcrbVtWLeY9CsX+1EyOIYCV4iK6xUO/Tigfi2dvcZf+VGypcggzI1m99S+oz+3m2V1uR2zFw+E++0GrjmJ5HZFip3XGH6UUO8FGGpahbeSsa5pcxp2f/UFTskxqPeC9Yq/N++CvpeNZOsG2RdqffquqV4QUW37qEpA1tQxtZ1bd/SzwrYTPH1BdLRsc7ba5avnawPr2V4E4U7C4WpxQJtS/h1j0Ces31tr+HMLYrHp89trkcGd5nYzrS63W0bW7ex+k0gOE1LsvMawdHW9Oi+UM+Y1FeIbtqQMCxwhBJWMhaIEN/WCfiM2ElNTKK23Ypgo2NiGim2qBLHAUraPTzm13Ga+7nKkZHNsJIupr49ff30vjDm70uULZ1cx1PR5ww1KdxOE2wmHvttxretRzphpnRxTH8yp3wy15YYIIbA0hbYb8eylOlPFzEBoLbXSisy2qVLJWNR6/rpA3VuRwd1Ezc2wutyOGVm3q/tNIjlsSLHzGmOni+9e7pg3d+XeLHDuHMvQ8WPytn7NF/RXc7c+U3GYKdtpxsk1HlsIwZlql3ovfQ9Hyw4rnWAQg7RtfEoY88yFOvP1Ht+cV3lgukAiFGxTZaJgg4D5hsty0123ngnKGZO8re9ZmA0LhyRJ+NrFBl95ZZVGLyAUcOdIjjffUeGuiVSompqCFyRcbnncUclQdgy+eqGOoaUusr7QWusELDZdxvN2+vllLRQl3FSk8Gpci6i50VaYA+ESug5uR/ebRHLYkGLngHEzzPRbL757vWPeakk5MZ69wiVyLXfcXhhjJslVn7tbBtBSy8X1YzRNYaacGaSS7/a++scTScKleg83iFEUhYfnyhwbzQ3cVEstl7VOsCmGRwzNpxfE1Hohq92A6aJDGAuAwfGOjWSoZC3uGHHwo+SqgeDbuZpeWGzzxLk1at2Qei9EV1WW2h7zdZe5kTSTLIgFtqkyVbDRdZWmF2Fo6sBN2RdaSZLQ7IastD2KtsG9kzm6QXJdAvVq7IcV5lotebeby0sikewfUuwcIG6VmX6vd8zbWYVslT2lZG/H0xdqFEfEkDvlyudeLQNorRPwjYUmlqaw0vaZHcngrMfk9AVLrRsOBIulq5tEUq0XAFe2UBi2lgzH8Fi6SiVrcrnlUXYMxnImGVMjb+tMFp1Bf6oTY1mCOOG5hSZfOL3CWM7k9UfL3H+ksKPQ6c+r72oqZ0xqXR9D06hkDe6fLuCFqZA6WnYGGXCGmrokIY25QUnT7PtuSlVVB722AGrdEMtQeGmpS90NGM9ZzI1krrnWz1YxMfz7flhhrqfG0u3k8pJIJPuHFDsHiBuxQVzP3ey13DFvZ5K3dJXxvEUQJYzn9x7w3HZj4p5POWMO3Clbn3u1DCA/jCnaOmGSWlWG33FfsCw2XTKGhqltiKRaN2Sx6TJVsCjYxkAI9OlbS6aLziCGx1YhiBLq3ZBwPUYnbxtESerSOzmebqizYYYgjHnyQp2LdZeFepfFhs5aNwQFHjhSvKKuTssNudzssdzyWen4HCmllW5HsjZ3TypUMhb3TuUIYoECmLq6KQ7HMhRGshub+lwlg1j/vPqfW8ExmCpmUBSXrKmz1vVoe3FaHLLhcrTkMDeS2RSDtFtl6r44K2dM7pvKc3alNxAXJ8az+xKYe601lm43l5dEItkfpNg5QLzagNvrvZu9IUGUYuPfXQVXJoNYXuaVapesmWViPVNpu81VCJG2VMhdmSnUn/NMxeFIyeZCzWWu4gyepygKQZSgqjBVsDaCjtfPsRCCjKFhGQp+KPjGQpPJgjM4Z/1AbkVRBl3W3SCdDwroqkoUC5691KDeCzlaDjix3on8Us1lqeUSJ4LZsoOuQDeIcQyNei/Aj5JNhQxPL3e43OxxrtqjG4ZkDYOSo1HOmNw7lSNMcoNz46zv131r2lon4HLLY7rooKrB+ntUuFhzN30PIBUAd45lCKKEta6HHwpsQ6XtQS+IeOZinUv1HkkCtqluOh9bP9O+1ezlpTZhIggjQScINyx0FYeZisNsxRlYjK5FiL9aF5R0eQ2RyUC1uvFYInkNIsXOAeLVio5Xcze73R3zXjeAfqNJL0xfP4gTGm64veBSFJTxce4YExwZOra9ZS/a2oPr4bnyYH5eGA/iWixdRVVVdDXd4Fc6wSDV+8Jaj5W2D8BEwR681onxLBMFi5yl0fFjvj7foO3FKMrmc9YPfj6/1uU/Pj2Priq8cbbMG2ZLjOVdXD/ia/NN3CBiumgRrFdq7rvOyhmdd79uCkNTOLPSo75uxRpOa+9/ZvVeRC+KGM/bjOVNLF2n1vU5u6JdcQ6HRaAQgqyppwHS65v61u/BTNnmzEoaiF3OmNR7AR0/IW9rvP5IkeW2z3zdpeNFtP2Iy02PqYKdWqmGXH/D4slad5OFicBQVdp+MAh4Hs9bVxRovBYhfiNcUNLlNYSiwNjYrZ6FRHJLkWLngPFqMjduZE2Pa9kAhl+37Ji7xuD0udr73Lxhp92+gSviWvop1B0/XndL2YMMpO06ffczsIbdLWkqvkvZSYXI8Hsv2jpfu9jg6Ys1sqbBeMHm206MMlV01lO7bXp+RDFj8vWFJuN5C3d9LhkjR87WCRO4fzrPC4uw3PIwNZW7JvP9Ez0QLRkjj2UodLyY5y/XMTV1IDi2WoL6IvCRYxVMTaHjRRjrn/emwoZ5izPVLl84s4KKyj1TOUZz9iCLrpgxKTgGE3mLpZZHte2TMTQ0VRm4JHcS0fdPF1BQNlXiDuK0iOJTF+qbxgN7FuI7vd5exffwuOt1ee1WS0kikdx+SLFziLiRNT2uxUo0/LppZpaKouwguHwffvqn08e/+Zs7VnPdTrgNByX3XTcAJccgiBLuHMniWBoTeXso1Xtzp+/N6fMuEwWLO8cy+FFMredzpqoyU3EGzU67XggIcqaBbWqDY9mGth7/opA1dTp+6sIJYw9NU5gq2FiGwguX29R7ARlD49Rym8Wmz0rb52jZ5tyqS63nM1GweeRYBUtXaXtpfRxT0wiThHLG3HQOrxCBlQxnVro8cyFtLNpvbtr/PIQQfPXcGo1eSN0NOVqx+faTOcKEwXGHiws+Mlfm/GqXC3WXJE7wwxhzBxGtqir3HylsEgW2moqN8dx6DFduI4Zrr0J8u89+r+L7eqw0W19vu7ID/XN/W4qfPa45ieQwI8XOTWa/YwNuVE2PvViJtr6X/uvuKriiCP7Nv0kff+xjO154txNuw/E2w64bQRqjMzfqcGwkO4gR2W4eG8dIs7GePF9jIm9Td4MNV1bFYTxvMV/rrVcitjlSchhbtwRtnd+masR5a5CRVXZM1roep5a7uEFIEAmEAgh46XKHJy7UBi075kayqKo6FEScBiVvLfq49XMRwHzdZaHhAgqjuS7lrMHousjww5hKxqKYMShnjIHrrl9ZeXMmXVr88Nn5JvN1l5cut3jgiMdsJbtj2vyO3zdl879b3Ye7rYHtPrvtMv62s75cjyt36+td4QasOFzaEgN1WwmePa45ieQwI8XOTeR2ig24mpXoRjQ4FEKw27vfepytAiOI09iVJy/U8cKElU6wqYP4dvWE+vVuxvMmf/7Nyyw2faaLHg8cKW1khBmp22u+7uIGMY6pcc9knnov5KmLjU3vdzuBBzAXZTE1hecXBW7QxDY0RnM6lZzFkaLDWtdDEYIgijaltl/tvF9xDqKE6aLFQt5CRXB+pcup5TZ3juWYLjk03DSG5ttPjFHv+fih4NlLdUayNvdOpbE0Yznziky6OElwQ0HbC1lupxv+Xr+rfpRQbacxXNW2P6jMvLWK83AA9NU++71YX4YF8bW6codfb+sxFPbugpNIJAcTKXZuIrdbOuxuomW393K1qsX9/zlT7XAiu33TyZ2OMzynvstkL5vbVnF2pGjiRQmJSFIBMZ5FUdVNVqqZcmawOX9zocVK2yfvGMCVzVO3nqt+jImhKpQyJoaq8Ia5UpoSLgSLDZdECExVZa3jcXq5MyiIeDWx2N/UB/WCgpjpgs1ax+fppQZCwFLLp+wY5J20/s7Dc2U6vsULi01OLXcJ4xbn1zrkbR0vSNA0BUSa0n7/dJGxnEkswDE1xnPWpgajO2XP9dnNBbmR8r/uYiykRQ+vlup+NetL//O4Ea7crceAvbngDnVGl0RymyPFzk3kRgYQ32q2vhdDhWYvIGdpm+qtbL1z96MEe/1xte0xM5SGPcxerWB73dyGN8ck6XFqqcXFeo+cofPg0SKsC4hhUdUvxPf1+cagRYauQMk2rmgUut1G189Sy9s6eVtnpuxwZqVLteWx1vUZzVm8tNRmrRey1g03FUTcemwvjDd1Se+nfi81XZaaHhMFmxeX26krjoTpkr0eHJ1Qdkwu1V2WWx4dL6bnh5i6xsVaj9FcGph8pJh+Kn4Yp1lUBYf7pvOECSAEXz1fY60TkCRJWmvIDXb8XLZan/pVoYdT/k0d2m7E1y7WCCIG7TZmKxuiZbvPeifry/BauhGu3K3HuNp37Hay2kokr0Wk2LmJ3MgA4lvN8HsxVPirF1c4s9phrpwhZ2t0/ITtrFfDm9L4NoKvLxoQYvOde8W5QpAMz+Vqm9vw5pgxNJ5bbKGiYOgKvSDmqQv1ba0WBccYZGtpimC+7lLrBpiGOuhNtXWj6x9DV/rFAtPMrLMrXf727BqGopCzdYqOQdExCOMkfR/bzFsIwamlNs9cXA9Anitz10QeU1Nw/ZjnF5r0wpisqeFHMZYOM+U8/8N9E7T8mHLG5MRYlqcuNmh5EQ3Xx1AV8rbKaC61BsVxwsVaj7KXio8kgamSw53jWTKWQZIkg2rSpqYCHh0/GQR49+N/tn4mW1PW+3E/hgrPXmrywuU1FCCIE2YrGRbqLvN1l5ly5qpiYfD9qziDEk/7ybVlDx58q61E8lpDip2bzI0KID4I9N9LsxdwZrXDWicAIXj0+AiqGm9rvRrewGbKzqa/bUqrzllpsC9sW7fl1dRdMVRY6wQoSirOukFE24vxwxgvjFjpBBwtO9w1kV+fp03RVvnMc8t8Y6FBzjSYKNiD3lRbM7yCKEnT792IC2tdNEVDUQQLDRddATeMefSOCifHcxwbzTJfdxnPW8RJghdEm9o2+FHCfMNloeEBgvH11xVCECUxl5upayiMYyxNIWMazI5kuGuyAEAYJYNsKj+IeGmpQ6MXMDeS5V33T/K1hTaOpXOp1qPlhZxe7gICL0o4U+1wcv0c9KtJ52xtvZ5OgBckfH2+sWPszU4CwAtjukHav2ux4ZIxNVpuiKGpuEEaiLydWNjOera1eOKtuoE4TFZbieQwIsWO5FWTt3VOjOaADidGczw0UxykNu+2+Tx9sc6Yrw42KS+MB405IY0z6W/sT12o0+yFBFHC0bKNOhRfs1eGheZj943zZreMqaXFCBcbDWqdhG/ON4gSQbWVYbbscLHu8syF+nrVYZ+MqWMZKgVbp+cFiETHMrTBRleyDZZaLm0v5HLTQ1VUYhL8KKHtRXT9iErWxNRUbFPnrok8YST45nyDz71UZTxv8bojJe4/Uhi8x6Mlh2rLAxj0xDpd7XK5kdbF8YKY8ysxc5UMhZzGA9NFTE3hM88vc7HW48RYnnfcO0bR0fji2TXCBNwwQVFVZsoZDE3B1FTOr3aYKtq0vICxvMXzC62B8JvIr9cvWrfQtL1ox2KMfbYLKu4XhJwsOARRgqVrlBydjJWWCmh427cN2c5NdJCsKYfJaiuRHEak2Dmg3E7Bjqqq8th94zzqVcjbOqqqou205zgO/qkzfH2+QUMYJO2NNOKLtd5gUx9uzNmv2zJf69F2Q75wehXH1HbN5tnKdufzUt1Nqwo7JiNZg2rHp+mGWLpGGCd4YTyU1g3jOZt7pgpEkeBvTlX5w6/0mCtn+HsPTnP3ZIHZisP51S4rbT9tPLru0stbJvWeT5gINC3tp7XSDTgWpe6r5bbHK2sdlho+l+pp1eeOH/Hm42U0TeOuyTwzFYcwSsiv1xSar7uUswZHKxm8MKLlRoSJoGjp3DmW4Rvzbf7m9CrRuovs0TsqjOVtvuX4CC8vt7l7Ij/odj5TtjE0hb98XgHaHC1nOD6SYaUb4AYx1bbPI3Nl5kazV7j3FCVNtxdCDNp09Bl2NSVCcLra2RTkPFtxNlns7hzL0PHjbbuwbxXC/e/MdtaUW7V2DqzV1nHg3LmNxxLJaxApdg4gBz3YcbvNRFVViuudt3dFVTFP3EEp38Fvb87Wqa5nOzmGxtxIZlOw8NxIhvmGS8sNOLvaZbqYpkLvpbpuP+6l3+zy5ESOFxbbfOHMKoaqcHIsw9lqh7OrXUazJpMFC0NVWW77g7RuXVV441yZibzF355e4dlLTS6u9Ti/6lLJWRwfzYGiUO3468UBQyxdIWcZOKbCqWqAEIK7J/I468UJ+8Xy4kQQxYJIxPS6Ia1eyMVal+WWx3senEJVVebrHkstl5JjoquChVqPBMG7HpggjAVnltuECUyXHMIE2n5A2TGouyGzlQw5SyOIEr795AiP3lEhZ2l87qVVzqym1ri3nhwhZ2tMFDOM5QweOlriYt0dtP3Y2hF9WMhcWOsNYp62+65erLlcqqdCdjibLU3xdwYZWTsFtidJ6lJbbrppvZ6xbNqjjCsDh1/N2rmdbjCuCVWFY8du9SwkkluKFDsHkBttnr+RF/HtNpP+nPd6/N0KBgKDzXUYaz0V/LIKpqalG/Meq+u6QcQT59ZYbvlUWx4ThbTFhKGqhHHMSjvk7GqXKEpTxkfyNoqqUG37lBxzUFDw5HiOMytdar0QTQFLV7ANFX19k0UIXD+1BsVJ6rp6YalN3jIYzZlkbYO33FFBVdVBlpJIElQFToxnCaOYphczX+sSxiZ/e2aF8bzFfVN5Ljd7nFru0PUj3CCi6YaYhsZDR4ocG81y13iexZbP0ZKTWl2KDg/PlcjbJg/PFa+osqwUrI04Kzq8+VgJPxRU2x6OoXK56dFwQyoZixPj25cHUBRlIPCavRA/jBnPmxQz5hVF/voWGWfd5WdqyqbA6wemi4PCjltbRDy/2BoI02MjGcIkra3U/6yH18b1rp2DfoNxMzi0Yk8iQYqdA8mt6nG1F151ddkggH/xL1AA+yMfASXdiPZSTO/EeBY/igGfSnZjE96uuu5wzZ+z1S7n13q4QcSRUno+0wwryJkGTdejkrWo9wJmKxlmKw5NLxr03fIiQdML6fipS6eSs3j7PRN4QYRj6bz5eIWLNZf5hstaN20F0fYiVtseDTem3g3QVDg2kkUBDBWeW2hRbbmEScJLSx1qXR+hKGQMhbxjoiiQJIK/fOEyL15uUc6a+GFMy424sNbGjwRHyjZfX2ix5qbv602zpUFmlEgES62A1U6AqSnUesGgyvJ4wWWm4myKs7LWO8ArAs6udKh1QyrZVLQEsdjUqHV4U7R0lfGcxaW1Li9d9qi2PN40V+HkRI4gFhhq2s5DCMHJ8RxzIxlMXaXlhryy2mGh7oICY3lr3S0WbPrO+1FCvRdgqAphIgaf006916537Ryk+J8bzvqaA+AjHwHzSgusFHuSw44UOweQ7YqaeWF8XXdcV1zEd0nh3gvXW112sEEGAcpv/Eb6n//qX2268PYtPFtryvQJYkHDDen4CaoaDjbh3TY4f70mzFTRxg1iXne0hG3qW1o96HzrnQr1XtoL6/nFFq87UuLeqRxnVhTCOM2WGu63dddEnpmyg6ooCOCpC3XcMF6PZzGxdZVvLjR5ZaXHsRGHJI556mKd5xab5EyNJy828IMY29RwdBVL05gombSCmNmRDEkCXT/g/JrHKysusxWbB6YLuH5AwdJpkVB0TAxNoRdEfO1iYyAY7hjLsNBMhUetG7DS9pmtZJgupBaroyUH29B468kRHj1eJr8ukEayNigtTE1DV7nC3eZHCYYKL17uDJp/9l1ZXzybpq+n9YNM2l5IoxcQJ2km10h2YwM9tdTm6Qs1llsemgqTBYeZSoaT47kriguamkLJMTk+khnU/jm70mOn3mvXGyh80LKpbqiVJQxhhzXX51CLPYkEKXYOPFu7dF9rQG6/mFu/A/aNTOHea3XZ4bvGSS3m5C5z3q6mTH+OO21Iu21ww1accsZMO3Wv/3241cNEweLJ8zXOrnQBhYmiz4koix/GaSG9MEEIwUzFYbbibBJiG1Wc08DqibzFl19ZGwjUKE5YbAYst30sXUFXFRYaHooiuH8qzboKwoQjZYeJRPDMxQZhnJAxNNwoptkLWTXgYk2n3osJhOANRwvcNZlnJGdTd0Pabsi51S6nqh0SMcJoxqCcNegGEd0gYrHp8i13jHLvVB5LV3lhsT0QLIV1t9O9UzkurHW5UOsykrN4yx2VQZHDU8ttLtXSAOGL9R7megT67EgGFAVbV8laab+yjhfx5VdqdP2YrKXx+qMlVDWNywqihEv1HgsNjyhJePBokW+7c3Qwh74FqV9I8cJal+cW0gy5o5UMqqpeVcz0A4X7x9iLYDgI2VTDa/Z61vyr4aCJvZ2QrjbJ9SLFzgFkWByUHINa16fWDdMMnz3ccW1X5K7fAfupC/VXffd2rdVlh+8aVeEPxM7W3ljb1pSpZDZZonazeG33XvayifWzi2YqGaotjygRTBctzlS7fOHMKs1eQNExuFDrDrKQ+rFKw/22hvtVHRvJcGI8z1g+5IGpAh0vpONH6LpC3tQRgKGrfOudY/TCkIYbY+gaSiJQFcFaJ8QqKRwtZRjPCUxDpePHuGFE149RVfAjQdtPA4jHciZ/c2oV29B4bqFJlAg0ReF1R4pcbrqYuka9F6AAL1xu84XTKygom75TQSywDJWpooNjaqjKRlzRMxfqvLLSJUrSTuZhklDJWOvvV/DG2TLjBXvwffXCBDeKGMub5CyNkm1wYS0Nunb9CE0RuLFAVRTMHdLML9V7LNTTNhUACw2PY6O5Qdf54fO/9bO9HrfMrcym2rrmd3PV7QcHSezJKtWS/UCKnQPI5tYGCR0v5nLL29Qwcq/P3xrsuR93b1fbJIbvGseGXnK4N1Y/wPdI0aZaSq0kR0r2tpao/nvZ6cK39aK5l01MUZTUjRKm5y6Mod7zUVGouxE5y+DsSofZkewgC2y7CsH9O/KRjM73PzRF1tKotgOePN/gnqk8D86UsHSV+VqPjh8TxBGL9YBOEKCrCifHHP42ErS8kJHQ4K0nRzBUlaJjcLnp8d+eu0wUC55bbLHQ9DhSynLfNFQyFqWMCUIQhDGXGj0sXedI2WGuMsrLy2lvrzPVLmtdj7VOQMsLmSk7iCQhSRQu1tK09zAW3DOZ48Jamh3l6FralqLlUnQMToxnmSxlUjffULf3bzsxmjY/XWhx90R6nh45ViZKFJbbqVutG8S4QUQxY3Js1KDeC/nSK7VNVZOHg5p1VWE8n6bGj+VScbUXC8h+uWX2y7KwuTCloJKxrohf2m8OitjbSchIV5vk1SDFzgFkWByUHAPwmS462KZ6RbDodpiaQtkxB+JmL+6e/WT4del0Bv8/X+8xczTGMrSNysl5i+99aHoQC/Pk+dq2Vq2tVYv7bQuA684WC2JBwwvxI0HDDdaDoBOOlh1aXsBKO8D1IkqjuUE21XLbGxQ7HM+bLLc96h2fv3zuMm4Qc89EHsdUWWp5lB2Tk2NZHMugFyScXlnjclPQ9QIabsz5tR5xNMJYzkJTVUZyJm85PsJCM6DW8zk2kuVNsyN8/VKdXhARJQFeJJgqpWn45YyBqggurfW4VPMYyZkcKdrMjWZZ6foEYcxSy6Xphqx1PIoZk0bP529Or3Kk6KRVn72Ith9S74QESczpapeeHxHGCUdKDjNlh4ePVbBNnTAebukBc5UMZ6pdnltsoigqb5gtcXw0x1MXG3hhQpQI3CDGNjR0VSVj6vhReEXV5OHMvJPjOWbKNmdXetTdgNPLHVBYz5TbYgEZikfbD7fMfloWts73as1WDxt7ETK3i6tNcjCRYucAMiwO+nevqrq3jsv98bWen/ZF2pI2fKPv3vZ6pzuwyAy99krb58Jaj9mRzOZNc70NQ78n02LTTZtHaptjcdLGkmnX7yfP15ipZJitZDZdNI+Wbc6udKn3gk3up+3mPHwxHc9biESgqipzFZtqRyNvGTTckOW2h1XVODGeHWQi1bqp62s8Z9HopcIoSGJeXG5xcizL+dUeZ0WPqZLDY/eO0w0idFXlhcUGKy2flW7AkZLDUrvHdDmDaficGMtjmTp1t7O+ocO33FnBNlRW2h4dP2K8YFPOGJRsg7PVLkmSsNT2sQwVP4xpuCFKrUe9E7DUcsmZGmdX2+n84oScqXG55bPc9LhvMsdio8di06Xthrz+aIkgTnAMjaKjM1F0OD6SpdoJqLbbjOcsxnLmutCzEDBwQ8ZJwpF22nKjL1zuHM3QDWJOLbVBgemiw9xIhsWmz3je2lGU++vZdm0/IoiS9Qy8ZJMFZDRr8PxCi04QMlVMrUQ3Wti/GsvC1dbJdjciV7upOUzsRcgcBFeb5PblQC+nOI758Ic/zPHjx3EchzvvvJN//a//deryWEcIwS/90i8xNTWF4zg89thjnD59+hbO+sbQFwf9gMxHjlV2vJPs33E+eb7GC4vt9XYF6UYXxNfWJrEf1Dl8jncb23/d08udPT1neP55O61arJAGORccfdOFLogFtqkyVbAHVq1hZioOr5vKc7nR48lzazx9vgbr1qyCozOeszi7ksbdvLyUnhcvjHecc/9i+sixCrOVDNVOaumouwETBZuMld4b9IKIS/UefpQwW3GIE1hu+3ztYoPxnMFbjpV402yRkazFA1NFXjddwjYUbF3hwmqHlU4qpmYrDhlTo+n6tNyAthth6RpjOYNHj4/wjnvHcMz0nORtbWAZ+qGHj/IT33Enf/+NM9w/XaCSNblQ6/L8YpPLTRdLV9CUtO7el86u8uS5OpahUO9FXGr0eOlyl24QEceCgqOz0g64sNYjEQJdhWYvTAs8WhrfcmyEnK1j6TpHSxnmRjJU2z4tN0rdfdH6+RPphnWkmLqcgjgtlmjq6ff34bkyqqbScEMikX72DS8kihk8f+v3pC90zq10eG6hyTfnm0RxwpFS+vlOrmdnvWm2xJlql//w9CW+eGaNy830s+mvoRttfdn6Pb0ae10nN3q+txPDa283i9lr+RxJXh0H2rLz67/+6/z+7/8+f/iHf8j999/PU089xY/8yI9QLBb5yZ/8SQA+9rGP8fGPf5w//MM/5Pjx43z4wx/mne98Jy+88AK2bd/id5Dyav38V7PGbPX3lzMmirJ9j6GrzfNazPTXdafrOIhvfpMLaz1y5fyggOCJ8eygku5w9lVae0W5oh3AqaU2l+o9ul7Es/MNGr0otRyxETCNEHz1fC0tHpgk6Xnh6qnyQghEkuD68bpVKce9kzmEgOWWy8tL7TSzrWQzW8mgawokaaPPT319kdVOwGTe5O8/OE3R0Tm/1mU0a1PveVyqu/zZ05eYKmX5rvvG6Lghz1xqYOsqxazGVN5gpRsSJaRWDBXuGHVouSHfXGjwjfk6d08WeGimyJ1jGV66nDBf6/LkhTpRknCp6fHmmRIZw+Cbiw3OrMSoisJYvkAla9DxBUVHJxZgGypumBDFMZqq88LlNl6UUMyYTBRsJooZTo5naXghbhhTdwO80KZkp3VzSo4xiK2Zb7jMjmQ4NprlgSNF2l6IY6ZBz7aRfu+rbR8vSjA0hYylU86Y1HsBXphQ7fjMRdlNgcfDQcphnDY01TWVuZEsx0Zzg/UUxqn4jOKEuhuSt8yrfu/3uia3jtvJsrDb8W55rInjwHPPbTw+oNzKmCHJ4edAi50vfelLfO/3fi/vfve7ATh27Bj//t//e7761a8C6QXmt37rt/iX//Jf8r3f+70A/Lt/9++YmJjgU5/6FO9973tv2dz73Cg//24X0xvl77/aRXnrHK7Lh66qKA88wJwQTK4fC9g20HSnzcULY565WOfCWo9uEGGqKjlLZyRroa4/r1+vZzxvEUQJE3l7kHa+05yHU9+jWKQ1YPIWtqnSDRIabkCzF/HyUoeVto+CwmzZ4Q0zJXpBxPnVLmeqbZpuyEwlS9Ex+OLZDmdXeuiq4MGZIi8sNPnrl1v4cZXLDZe33lnm+EiOy3UXR1P4+mKbizWPSk4nDGOyjkHHizm70mah6dLsBjx9ocYTr+RQhGB+vdlo2w/IWAYzZZNK3qJbcyk5Fn4craeRj3BsNMtXz9UxNJV6J8Q2035kOccgYyiEccLRUoaxvM233THKyfHsoHL1UsvF9WM+8/wyQgjuny5iaArVljfod5U2Hc0yvl5/aaKwcX63xuHMVtLP9Ey1u23NnEGQcpgGKU8UHAxNYaaSueLOPm/rnBhLu7PPVjI8OFO4qoDZS3A7XBn/td2GvNPxtiv/cDNiTa64Vqgq3H//9T9fIjkEHGix863f+q3823/7bzl16hR33XUXX//61/nbv/1bfvM3fxOAc+fOsbS0xGOPPTZ4TrFY5NFHH+XLX/7yjmLH93183x/83mq19u093Ii7uuGL6XjOYm69r1D/QnSj/P1pATfjisDmrXMYvqBfrw99eNPYrQLytnd7QhDGCaoKIxmT8lgWRVF58/Ey1lCmVn+D1jUF09ioyHvnWIaJvImxZdPx12vAXFjroaowmbdxsmnatKFC0TFwoxhNFVxuuFi6ytmVHifGs8w30po8Sy2PvGnghxEX1lLX0nI7wNIVXjddJEwUzq31APjcS8toisJE3qLW9clnLKrtgKYX0HADVLXGQ0dLrHZ8dE1DVxSCGNxOyHOXLxOEaQyLqWrcfzTPtx0rMVnK0gliDC2Ns1EUlUfvqFDMWtimznytx1qrhxfFKErCSMbEMVUMTeVi3aVg6/zd+yZwTJ2nLjYG4nmiYPHU+dqgCnM5m/apstdT1G1TZbnlEYTJoC7OkbKzaePc7ruy3f8lSYLrhxRtfVB5uV/AcWuhSUj7sr3j3jEemi0ykjXRtJ0Fev9z3ktw+2zF2TGrcfiY263x4Uy98bzFTPnK+kz7wXZlJ67lxkemd0sOKwda7PziL/4irVaLe+65B03TiOOYj3zkI7zvfe8DYGlpCYCJiYlNz5uYmBj8bTs++tGP8iu/8iv7N/EhbkQGwXDWz3ytx3zD3ZSqC9dmAt5uA+gXL6z3gm37IW1Oh+9RyuiM5ixUVb028RYE8L/+r+nj/+V/AdPcFGxcdsxd0+uFEFysueiqwlTB4eFjJY6NZNPCdusbSV881bohi02XqYKdbsRRQrXlEcYJtW6Aoii8cbbMnWMZar2Qkq3hBjF+FFN0DEayBvO1Hi8ttnhu0abk6BQsjYKlE8QCTRGsdlyOlm0m8jYqCQhB0w1peCENL3UrOqbGeNagkjdxDNBUcMOYbhBxqd6h46cFBJtdn4xtoikKOUejZGkUbY2smSVrqTwyW+CVao+/fWWV5ZbAC2I6QYSuhpxaEpwcy6EoKrECbTdCQeG+qQInx9PviaEpnFlu81++sUQiBHdPZjlWyRDGKqdXWqy0fLK2wbfdMcpqJ6C9Pv/ZkQwFJ+2wXm37JElCGAk6ftrc9GjZoZwxKWdMllsei00fgeD8apcwEtTdgJJjpJ3szc0dzbd+b+M45i++ucRTl+oULYN33DvOyfEcqrrzuhFCDJqIprFQG2J5u417+PvmBQnPXqozkrW5Y2xzcPtMxbnie7mdGNhujQ/WrLu+Zusba3Y/2Srk/Cim2exy7x/8btr+41/8i20rKG/3fJneLTlMHGix8yd/8if80R/9EX/8x3/M/fffz7PPPstP/dRPMT09zfvf//7rPu6HPvQhfvqnf3rwe6vVYmZm5kZM+QpuRAZB/2IaRAnt9fiJ4VTdayFJkk3Vc7fWNml7MYoSXJHi3p9DkvQ4V+3x/GKLE2N5HrtvfNeN6ArCEPpC8+d+Dsy0cm6/71Wt53Omqu54R+lHaXxHwTGx1wvgbd1AN+aaYKgZcuvBvd+cb3Cx7tLzY2wdVFWjktF5/FSVl5fbFCyDckZnNGcyUbBYqHd56mILP4q5M8gSxWBo4IYJYRhzfrXLZKHFYsMD0orMiy2PjhsSJQmOrjJddkiEykTeZK6c5WyuQ9E2sHWVUsZgIm+z1G7RciNiofOtJ/Lcvf7e05gXyBiCpUaaIWUZKneMZmi5ASCIk4RYpJ/f46dXePJCk3LOZDxn0gti1roBpq5w92SBnh/ytfkGa90QVRXUehGnV3pEQrDW8QkT0FWVU9UWa92IbhjzLccqGGp63k+O55gtOzy/0OKLr6xiayotP0JBYOpF3npiBENTWKi71HoBy02PpaaHG4QstgNOjuZ49M6RTVWxhxFC8OylJp87tcJy06WStbi36XHXZGFXS+VOwmInywykwe3jeZOvzzc4tdwlTNoIMcp4zgLSgHl7PZZs+Hs5s37MfrmBmYqDM9R+ZKuLN4gS2u6V6fWvlp1cTcPCq+yY1Ho+vbbHyP/r/5EO+Pmf31XsyPRuyWHlQIudn/u5n+MXf/EXB+6o173udVy4cIGPfvSjvP/972dychKA5eVlpqamBs9bXl7moYce2vG4lmVhWda+zn2YVxt41xdMMxWHi7XeIEbiWi9EmztIp8+9srbJ7j2HShmd5xdbrHYDorjFo8fLFLOv/lwGsaDeC6h1w03z2oqlp40n/TAmjgXfWGhuEm39ufY3KSF8sqbOkaLJNxdAVxXKGZ0wjmm5IfVuyMvLLc6vurhhzIMzBXpB2nl8vuZS74VkbZ3JvEU3FPS8tC3DYsNF0xQUBMcnCmgKaTzPcpu2F1FxdDKmzkTB4dhoho6f0PIiyo7JzEgGYz3L7vhYlq8vtklEQsePudzwuHe6jG2otP2Ii7Uul5suXzm7RiIEE3mL0ZzNSNZmZjTLStOl4UWEUYIbJjQ9Fz+KsBTBWi8ijGP+r+eWWKy75G2NIIoxdVAVjemCRdsPWWz4FG2NB6eLHK04KKpCIhJ0RUFRBC9cbtNw04D3o2WbU9U2S00fBUHLDwkiQS8UfNsdFYIoYSxvkoiYes+n2g6ouwE9PyaMEqbLzqC0wFb8KKHjh4w4Bh0/ZK6S4dhQtez+d3jrJt93vwZhTBsGwmLYMlNyjCssM+N5i7Jj4oZNLC0tfPhtd44wN5rd5KJquOH6TUBay2c8ZzFf69H2Qi7WegPxtrVFRb9v2HBhzP2u93Nl2QqVFbe352PL9G7JYeVAi51er3eF1UDT0vorAMePH2dycpK/+qu/GoibVqvFE088wY//+I/f7OnuK+mdvs5dE3nmRrLXdSHa3EE6Lfd/LQUHFUVhNGdx51iO1U6NRAguN71BX6NXg6kpO9bUGUYIQRAlROuuqEJippvQujhKkoSWGyKShKWmy7mVLkttn2OVDOMFi/GCzWjW4C+fX6Lacrlkq8xUMiy1fCpZg7JtoKASxwntIKGcMzlacvh7r5+i5Sc8v9Dg6fOr+FFM6MG5tR6aplJwTNwgpNkL6IZx2kjTUCllTTKmTjcIaAURYSK4bzKPZWi87a4x3CBOs5oaLgkxTS8ma6mM5h2Mrk8YJHzpVJXllkciIEwEcyMZdFVjrpzlW+8YBQE9P+RzL1epd0N0VaHrR6x2fBpeyNmVHqerbUZyJiMZizfOlMhZOvdM5vjGQhtDVUgEjOZN/u59Y6x0Ij7/chU3ivHDhOWmi7ee9j+eM4gSQTlrUHEMaj2f+Uaapn5mpcsXz6wCgvl6j5eXOuQsnVLGZDxvkTF1xvMWCJG2Ctkm0H6qmOGtd43yTlPnvqk8jmVscrVuVzCy734dz9scKTusdIJNlpkgStYtM91N1p5+RlnR1ql1A6otj4t1d/d+bIbG3EiG+UYqjqttf1NdqK1WU9vUuWsyz1x0fWt2O67mahq+uTo5kWPWvrbyEzIrSnIYOdBi5z3veQ8f+chHmJ2d5f777+drX/sav/mbv8k//sf/GEgX5U/91E/xa7/2a5w8eXKQej49Pc33fd/33drJ7xOv5kK0tSnmfdOb3Ql7Obaqqnz7yVEEEEYJK92AY1Hyqi+O29XU2eq6EELwwmKbL5xdxVAVcqa+qTN3kiT8X88t8cVX1ojjmHLGouWHhFFa6Xe67PDwXJmeH/DScodLaz3qbsgPv/koU8UjjGYNRnM2C40eT5yvcaRogqIwmjNZ7UbcP13gaNHiC6eqnF3tIgR4UcJ8wyPvJZwYz1Bw0mBiRVFoeRGGkgZTizih2QtouiG1ns+dY3lA4WvzTbKmhmmoJAlcbnqYmsY9E1mevRhRbXZo9HwUJRWEI45BFAuEkqApcKTkcHI8x7mVDi8ttdIU8Y7PaidBKApGoiFI51hte0wWbO47UqBo62ldH0PBNqAbxjx+apm1jsdYwcbWVY4ULVpeSN2NMDSFE2NZLjc9hEiDtx85XgYBF+ouU3mLb8zXeepCjbYb0QtCOl5Eo6eRzxjcPZHn4WNlBPDV8zUmC862wbPDjVZhc/FHP4y5VO/hhmlxntmRDMCQ+zXkTbMlJgsb34e1bsBa16PjJyjKhrVn2M1TcAwavRBdT4Oshy1P290E9DPU+qLL0tVdrabXu2b34qq6mrXoRgiX/czOkplfkpvFgRY7v/M7v8OHP/xhfuInfoJqtcr09DT/5J/8E37pl35pMObnf/7n6Xa7fOADH6DRaPDWt76Vz3zmMwemxs5euBkLfmvDylfzWo6pc8dobtPF/tWyU02d4fm33JDVjgtC0AtiHj1e4eREfrChNHsBX3pljecWWsRxxCPHKxwfyaGpYOoa00WbM9UO1ZYLiQBF0PZCztd9Hpwp8dBMmbytp20HgFeqHV6stml7IZ9/aZkgSjgxlmVuLMeRlR7NbgCqII4FSRxTsHS+855xvjHfpBukRR1rXsS5WhrH0/FTEZCgEIQNSBJWuiEtL6SY0YkjgR9GvHS5wUrb5S9fWOZUtYPnx4xlDcbzDlMlm9VeSJzEJALafkzXj3lxscE355ssNnqgQtEyOTriMDuSRSTwwmKLy00XPxIoqoKpqJyvd1HXz22zF3DJTzhfcyk4OiXH4sGZEkdLNpapU8mYTOQt/uK5JRYbLtMleyAKpssZdBWeulgniAVBnOBGoGugqAJNSYOyw0jw5XMrOOvuniBK0oKNW/qK9X8/vdxhvuFytJRaSS6s9ai20viok+O5QdPVfpzNeM7iYq3H1y420u97kNCLIjKGztF195NtaFe4eYQQNLsRq22fkn1lgPxWwbCdAPLCeEer6Vb2st736qq61nW8tfnuXsbvV3bW1bJMJZIbyYEWO/l8nt/6rd/it37rt3YcoygKv/qrv8qv/uqv3ryJ3UBuRqrn9b7GThfl/fDrb401GK514ocxF9Z6LLVczq300qwhS8NQN1dTtXSVcsakYOsoisZkweFbjpcpZXRA8NT5Jl94pUbR0hgrWPTCiIJtoipQyaYFDc9Uuyy1XKJEpMHGXsRKyyNvp/VkvvfBae4czfNiqcNkzkJF4MUJOVtnNO/wuukCtq7xzcUWUZLQdSNMLbWctLyAcD3Aumbq1DoBfpzGhHhhTIJCxtD4yrka5YzJxTWXrheTtXTuGMtx35jDk5fSgGjHUDlacmh4EU+cW+XFxRbn1rq03RjTVMlZCm8/OcaRSo5ax2W17bPWCwljwXytR0IqGDOGjqapCEUlFgIvTNCUmKyVIOKI5abLfNPnvokCbziaZ7nlcbnl4pgaIkl4fjF1IU0UbL7lWJmXLrc4Xe1QcAzKjkarF1LrBry83CaMEi43XRxT5w2zJda6acacH8aDvmL9LKKio/H0hRqLTZ9qy2OiYFHt+ORtA9tQGc+ZnFpus9DwOFKyeXiujAJ88ewaC420UnatF5I3NVQ1YLRggbLuBh2yJp2cyDGRNwmjhLYf7bn/3FYBdDWraZ+dXHG7pcdfzVV1NYYFznDz3b2wn9lZe8kylUhuFAda7LwWuBmpntfzGnup7XMt8xwIp13uLPtZLMOviwLzdZdqy8M2NFp+iK6pCFL3ydxoFlVV0+BT4DvvHiVvalTbHioJXzqzynNLbaJIoGkCUKgnCfeMZ5nKO+g6fOud49w/XSAYamy50vK4VOux1vGJk4R6x+dy00VR4AcemuY77xnj3EqHuhumLRbciJcuN3lltcsbjhaYrTi8vNRhsdHj7vEsMyWLJI4JtZhuoLHSCai7EQVbpRvGRFFCEidYuoIbxMyWDI6UHRxb5/7xLHMjWda6PotNlzBOyFoqjqFSa7u4nk4UJ2Qtg16UwHp21lcuNEnO1YkSgRdETBRMwiBmuePjx2AqgiRJX3M0a1DJGmR0jVAkNN2IL52r44chQqhcbnokJIRxzGjOZCxn8uJShy+dXcM2NIIo4Y0zaZp7lCRkTI2/c3KML51dY6Hh0fUTzq526foJJ0Yd7hhx+Or5Bt9caJK3NEazBgVTwwsiul7IE6+ssdz0Bq0V+kHISZLgh4KnL9R5bqGFG8Ucq2SZq2QQisKRkk215ZAkCeO5mF4YD75fyy1vkAo/LDSWWj4rnQDYXAjxWtjLDUDfOtlv59JvXnqx5u6YHr/VVXU9VmA/Sujbuattj5lrcDvvZ3bWjcwylUiuhhQ7t5ibkep5Pa9xI++6NgVuZk1OPvFEeoxtXI3DwiyI0kD0foxG3ta5Z7LAWsen2vZZbbl87qUqOUvDDwWrHY9eELPS9Vhsuay5IV4QMV/30DSF6YLD8bEMd43n0TSFM9UOc6M57plM73QNVZAxNFwt7SmmoKQB8UIQkxAGMV034Itnayy1PJZaHsstj6yhYegKi35MO4hpdD3uGs+hq4KLdZ+VlsdY0SaMExpuTJIIVARxFFOwDI4WbBabLg03wlQUDA1MQ3DnWJZjiWCyYJG3DZ65VE9jpRKIBXztfI1QKJi6ytFKlhOjDkIkREKhYOnUOy4NL6bWCchZOjlLwzZ0LCPC0BVGMgamqaEKQTlr8ejxMpMFh//y7AJPd1vUOh4CQZIoJMAzFxsYukbB1rENlZeWmiw2XfwwxlSh0Qu4WOvRC2Jydtpy4i13jvL85RaBH/DsQpPLjQA/ijlazlBth2iKYLHh8p+emafomJRsnQu1Hk0vomgbfOuJEe4Yy3Gp7lHvBeRMgzDyaXkhF2rd1Aqjq5xa7tDyI8ZzFt/74BRnV3qsdT1ylkHWSsVl1tQH8Tt90Q8MLEaOqTFbdvDCGAWu2aWyXUbWcKHE5xdb1Lo+fijI22msmWD79iXbiafrtdBauQwX/uLz1LoBY6Ola7rG7Gd21o3KMpVI9oIUO7eY67mYXOvd3V5fY/i4W++6Wm7AkqZc813XFYGbkzlm3/CmHY9hagplxxxUcRYIFhpeGiuznoXWdkO+er7G6WqHry+uMlPsj3Pp+Wk/qYxh4Jgq5fVMG8vQedcDE7zlxCiWrvJ/PnuZbhBzptrG0BSOjWS4uOZyZqWNoapkTZWjZRvH0OgFAfWuT5DAUiug1lvFjwRhIgjCGFWBsbyFF6Rd2l9ZdclbOlECbS+gnSQstDw0VaFgGSQCRGqAwdQ0/sc3TfP//vw5ar2Iei9at85o3D2VI0rS81fK6Kx1QnKGBgl03YClQGCoMF5MLRnTIzlyjkEnELzxSI6FVsjl8zUEUO8FZAwbQcJkwUGg8IajeXpBzKlqlzBORY1jKFS7AW4QESUJcSLQVZWModJwIwwtwo9iXqm210WVhqurNHsBf3tmlSBO0FWFybzFSifgzXMlRjIGXzy9yko7QAEWmy6X2z6mpqEpaYNSlYRaN2Q0Z7DYdEEo5CyNN84UKWZtnjxfY60TcMnvEguFJEkYy5kIoJI1We34eOtNSSeLNg0vpBsINC3hvukCUcIVQqO/qfbbWIznLS6s9fjapQYAb5wrDzKzrqWX1nZuqhcW2+trQOGuiRwPHi0NqjbvdCOy1Xp6vVZgRdeZfefbmLhGi9Dwe97L62xttbGX83Ujskwlkr0gxc4Q3S5o26xpTdtshOh2dz6Gqm7utbeXsf2LWq+XdqreDkWBTGbjYnp+2aXomNw3VbgiPb8/to/rQpIogEYvuPLY2ezGcS9UXYq2xb1TeaZzOSqmzRdba7yy3EWLNqrI+lGCiNT1425PNruR7k6o0okENhaRr9LdMo90vmkV56WGT940mcpmObvaxe2BaynEgYYbKuQdk6lihtPVDnnNJArS4oLVeoylJFQcg3un8ozmbXqxR842iJKEkm2jC5Oem+B5CqGnMN/yabUbnMm6hEpE0w3o+BFvOFpmwslC7JHXTHxPZdTRWW75hHGMoifcOWYTxwm+r6DGFidHstTbdZqdiNVmTNEyGDEcztRaRCLB0CGralRsi8iDlidYrif8zcst5ldCXE8DPaEbRFQ7LuMdm3PLPVY6Po6q0AxigjjBVBVaPvgiwbEElq5QtEyCnkLk6+RUFUu1CdwQ31NodBJMPabWcTnmFDhWdvB8hXo7IWuYlIyYpbrHf/7qZU5O5MgbBkeLFg1PJY5AFTrjjkPGTDukN9yEVSutojw7YnHfVIaz1TYrTY9WD+ZGM6iRyjNn25xb9Ki5AYtrPeIQVB2mizZHCjYFM0tOMVESnQurPRIfWrEg9hVWuy4CwdcuNXnHPTaNhuDMsksiBMdGHE5XXequoOzo3DtWpJDXWekEjOds1Fgnr9n0EhdLmPS6sLgaUOtFFB2V1x8vUsyYgEKvB9O5HCN2BiEEX3lljXNLPiAoWe5gA07XnMd43ubE+JUWlf41oi9Iqmsxbs9jxE4X4sKqT+JrdJKEjJq2p+gfo//6lq7S621thbFxPbF0lYJm4wqPvGZfsY62rvvN15PN63/7a8TG72lF9Q7VdvqeHzy+8Z63jt06fm7CApFazEqGzfHRnS1Q2Wx/PgrE21+f+mQy6bwBfB+i6MaMdZz0PENa4D0Mb8xY297YT65lbBim43fCskDXr31sFKXnYidMEwzj2sfGMXjezmMNY6OO5bWMTZL0u7aXsXtCSESz2RSAgKZILw+bf971rs3jM5krx/R/3va2zWNHR3ce+/DDm8fOze089r770jFuEIm/eXlZTB3zdxw7N7f5uA8/vPNxR0c3H/euh3o7jrWdRPT8ULx8uSUeP1UVb3tHuOPY/jcrSRLx8uWWeNs73V3HdjrpHB4/VRXv+J7urmOXlxPR9QLxjYt18X3/cPexv/GfLolf+T+fEx/59HPiHT9U23XsL//v58Uvf+ob4hf/09fEe96/suvYh//ZE+KffvJL4j2//Tfinu85s+vYO//Rl8Rdv/hpcewXPi2OfPdzu46d/MGvitf/y0+Lt330L8VD/+CFXcce/YGnxLd95C/FT//7p8T/9KGLu46d+nvPijf+yp+Ld/7m58Tf+b/vPoc3/4Oz4od+73HxDz/xBfHtP/nsrmNf//2viD998pz4wy+cFo/+8yd3HXvifzgr3veJL4rf+PNviv/vn6/tOvb42y+KH/nfvyJ++7Mviye/sft3593/Y1u8dLkpen4olpeTXcc+8ncb4oX5unD9ULTbu499598LRJIkg+/lbmP714j+9912dj72t397LJIkuc5rxM7H7V8j+tx3387HvZZrRLEcCzeIBmPf9radx1p2LP77C0viv7+wJD799UXxLW/zdj1vw/zgD+48DtJrRJ/3v3/3sdXqxtif+Indx547tzH2Z39297HPPbcx9pd/efexX/3qxtiPfWz3sZ///MbY3/3d3cd++tMbY//gD3Yf+yd/sjH2T/5k97F/8AcbYz/96d3H/u7vboz9/Od3H/uxj22M/epXdx/7y7+8Mfa553Yf+7M/m47r79/NZlPshrTs3Gb0M46EEDuOEQiEYI/m4DQDx9QUyhmTJNn5uMr6T9+U7kcxVzMO9l1olezVZ9J3nRna7j77V6pdwqbLeM4auCB2oujolPIOHS9Evcr5mCrZKMWQes+n5ce7jvWCmPM1l6yhE0bJrmMdQ8MTIAA/2vn8AhgqeCEsdUI6W81fW9AVhZ4f8uLlFpZf2XWsENB1ExAewt3lFhMIk4RLDY9KRsMydv8sOl7El862UBQ1fZFdyDsWjq3T9AX1WgvYec6OoTFdsrl7Ik/R3t2FEiViUNzvaiRC8OSFOostnxErA+R3nq+tbwkWvjr977u6y1et7UecXvavM+tot/HiKn+/PlSFPcfRKIrC0bJD37Kzf/E3+/NeJYcXRey2a75GaLVaFItFFhebFAqFK/6+326sPntxY0Ea7Pj02Ta1rs9EwRmY1YVITckrHY+58Y0Axu3MzsBgfDtOM58SkfDFl5uowH3TRe6fTl1kQmz44rPZjeaKJcPm2MjOXZWzQ3uP50EUCfxmG/voemuP5Spkswgh0MwEe31jbXbiNMW87TOWt5gpbwRtemHM89UabS+i4Oi8fqqChsqZaofllosfCCxTYTxvgwKt0E9TixPBpVWXr51vcn61x2rHZyxn8fBsmeNjGdp+zLEJi+cWWyw0XRqtiBcWWqx0AgxN4UjB5HLbJ4wE3TChnFfJOzpFW8fzFVQBhq4ThRGXmi6OrlDOOiRCMFKAl5baLLYiolhBj1UMFQxTZaZo8x33jPLycpfnLrdoej6hEKhAEisksUoM2MB4yeAd94xRzpj87dk1Tq+16AYJGVNhtpRjKufw4lKLUMC3HK/w1jtG+Mq5Gt+cb7DQ6tEVaSbcqKGgKHrad8vUUJVUhOUsheMjWc43ejR9nzAWVDImpmJQztjcP50ja5q8sNTg/EqHpp+Qs+FNx0t0g5iLqy4aOt96R4UoEbx0uU3dDZksmDwwXeKOcYfFjssrq12KtomWaDR6AX4gMHSBaWgkAhxTY7pk8dDxEm+cLdFyY/4/nz/PC5c72LrCO+4dp9YJuFjvkrV0Hporcedkhply2prhK6eaIKCcMVjr+sSJYCxn0XAjEiXBsaGQMSnYBveNlblU6/HU+ToXal0m8zYnJ3O8aa6CrisUc+n3OgxDXpx3mSnZJIp2xfd96zWi07kyxscLY56+UKPjR4yVdR45VsE2tFd1jeiv4eWWSzlr8vCdG27twdhuFybGN625q7mx+sfuv4dcbuO99noCN7gyHqc/vlLciNnxXMFXz9VpuzF5R+NNc5VN8T9brxHxLvcYw66pZifmibO1HY8r3VhXjj2sbqz+/t1sbr9/95GWnSGy2c2Lb7dx13LMvTJ88dkNVVV5+ERhm4tpQjv2CIg2pXAOXyyH6Y8fznyaqOjUuwGd2ONyVxsIphwbF5LNRdm6e8oOMc2E06stVpZqfCfrvXqyIDJXBnTatkIn9gmUkJqfcG82g22mx80IlUnPRlHS8fmMkgZQRy6RGlMsqmm8Ts5Kq+5GWQw1tUAcKVo0PJ9m4NFLBKWiSjtxeXKhQ9uLWe7Z2LrOZMHivqkclZzKC0sden7AxWaXbhChqwrHxm0c0+DEWI6posXFhkutk2YWJYpGJudg6BozRZtAQL0bMFG0qXU7dBHEWkwMJEnMqhfx7GJqcbDMhAzg+aCoECmCREszgxIgAF5ptHhdtkisRYQiGVzQ8xmVdtijmwREQtAOXRI9YrnXoRm5+Os7pABqoSBjhmiGgm6KtL6iCj0h8JUI01TQQkhEKi6FAZ04oe7raGZMO3CJ1QjHUZgZsSlnDRxDI4wTLE0h0WJavoebeGh6gmYalIoqR0ZtHEdFJIKmH9KNPJZ6LvN1j7G8wb1TRR67Z5xaLwBFwYsSwgSKGZ3Xz+XoxD5FR2duzEjjwZSQuyZy/J2TFS7VPb78So2Feo+19av05EgeJUhodn2sGL7v4UksU+dS3R1k/uSysHa5x5rfJVIDXBQmKyXW/B7V1XTMsYrFb/zlaV5YbnGk6PBdD0xypJzd9H0XYsNCGkQJC+2Nflj9cRmhcmzCvqIYZyZzpTAaFhrDFoyt1wgvTGhFLhdbbc42BBlH4f4j6QVf0WNUwMoIlKE1xzbXpO2vEZvXfX9e862dMsI2j0977olN77lSVNnJmHUtdWALWXXPx7Ws9GcvmObe40D2a6xhbAiJGzlW1zeEz40cq2l73+euZayqXtv+eTWk2DmgDF/sdupTtTVDYrcU880Xz/Suy9SUwfjhmjYdL8ILkx1rXvRf21uvi3G17BAhBM8vtPizZxeIWm2+c+hvfpSw1HKpdUOSJGGikBb3G8+nzRZbvYDTyx3uP5LesV7Z6LDL5WaPznpPKT8UvHi5RSVrcf90AVNT+O8vVDmz2uHOkSxvmC0xnrfohQm6IvjapQZfPrtGJ4j52iWNeyazTJeyHC05fOe9E7ztnjH+f09e4sxql0SAoRncMZpjtpzBi2JA4OgaQeiRsXSiMMaNBE0voNULGM9b1DoB3SBB18BIIFy/M4+Bji94aaVLztRRFMEdZZvVbkiigOdHBALCOL3L6QYx51Z7NN00kFpTNUpZhTvGHJJEMN9y6XqpADq72uXzLy5zbrVDq5cwfGMrgChOqxonAmxDJ3QDEgELDY+Zkk0UR7TcGBSBoSs4usZq16Pjh9S9CFSVrK6QCBUvgvG8TdsLCQWcrrapr3cFdwyFpabHV15ZY6UdULJVXlnt4Rgqhq4RxTG6kma2TeQMat2QZxdaFCydt9+dIYnT8zmSNXjzXImFhsvjZ1ZRFIWZSgbH1AgT1sskBGhKOhdDUzhSzlDvpc1e3TBhvunzwBF7U+aPEGkFbC+K6bgxuhpwca1HIWPQ8ROEcLmw1uGLr6zRDdM0/vuOFNB1bfB9j+OYr19q0fJ8gig9r9V2ms4uhGCisBGQvJd0cmDPKeZ9t3aYCAxVpdbzB0U4n7lYB+CNowZ3b/vsa+daM8L2K3VdNiyVXCtS7BxAtrsAXm0x98XMdu0gho83nrMG6dxHy2lfpZmKgwKYuspcJbNrl+Zh0WTpKuN5Ky3Zn9+5PH7/AukFMUGw2U7dbwC6UO+y0kxrlEyXssyWHS7Veryy0uULZ1dBgQeOFFEUZZPYWmq5nFrq0A1CHjpaxtAEp5Y7hEkbBYWjJYuXllo0vNSG/OjxMtOlDDlLo+2GLNR7mLpG4sZESlpMz9Z9vjnfxDQ13ni0TMbUKTsGtV6ErQvafsw3FpsstXzCKMExNNpeCKrCVMGm6Qa03AgvFKx0fMIopuNFkEDJAjeCIFoXHQJ6fkIcReQcnVaYYBgaUaJg6gIRxTiGgkDQ8KDphggR0/UTNEVhumBzz1SRJ8/V6XkxIaAKaHQDLtS6eGGCUCCjgKGn8UCRSMWOqkEQCUbzKkGs4oaCei9tJGppCo6homoqU0WbbhBzodbDDxIsQ2M0ZzFRsMlndEYzJjEJGdvg3EoXgVhPWYemmxALQa/aYbXto6gq9W7A7EiGN84Uece9E3zhdBU3TPujXax1SZKEatvlq+fX+KuXl7nc8NKu6wU7bYg6FFeWNQ0cnbTdRa3LHaM5vvOeMTRNw1z/Pi61fbK6wny9x4mxLI5lYOlqmlEoBI6lcWIsR9NtoCkqlxouj5YzqGpM2TGptroUbY2OF3JkNMtMyRmsjTiO+S9fv8wXz6yQNXXG8iZHyqn5xTZUvCDh6/ONQWPQvaSTw/a1d7Zd82HMHaMZYJR6L1gvKwDzDZeFhocQCaIjBmJHiGtrF7GV7W6otrsx2/p/+1EkUDYslVwLUuwcQK717ulq4mj4eH4Yc7nhpemxLS8VFXVv03P7XZqH2zZsdxd6Yjyb7tgAYmdrlKWnncVfd6SIKG185ZIkYbXjY2iCWMCLy03CJEFVVWYqDpMFh1PVDoaqUO8FaSXYLSX6S47JYtNlqenR6IW8/mgpTc/WNFY7Lm4QEicJUZRwfMRJq+V2g4Hoy9omj8yW+PpCk+W2z1LTxQtiXl7uMDea4Y6KTc4xODmeo9YLEIlgqemz1vWI4tQioasakMa8zNc6uGEyqPnS7MW0A0EcQ1aHYtZE8WMUP8bQVESS4IrUHVGwE3RFIxAKvTCk1Y5BBUsXFByTohITC4jimE4AuipY6UY8faFJ3QvxBZgqxEkaq9H0E1BUbFOQ0RRG8iaXmx5uIPBj0IRAVVXiWKBrOonrgYDVtkchY5OGuoOlqfhKTBKnv2dNlZlRh0dmSiw0PM7VuigoRHFE3jYwDYXXTdmsdDzOrnTTz1RJ2360fR9T1wmiBF3TmSo4TJdytLyQ82tdTo4XiAW0vIjnFpqstgNWez6WruPoCqWMQc+P0TUFXYHzax0+/Y2Ypy+mlqO6GzI3muV1R0tpEL+AiqNzutphab3Oz1vvrPDScpeOHzKSs9PvghBEEXTDkLlKlgePprV5EiFIRMKbj48QxAkPzxZ54GiJ0fXnfPWVOo+fWmax7mEYKpOFUSpZk8miw0Te4hsLzfUmpduv452ssVcrAiqE4NRym2cupNabN8yUePOxyqCB6tGSQ7XlEcYJerARJDFcTXkvbF3TWy0qcKUVarv/20kESSQ3Cyl2DhD9C8Gwe2kvFUW9fjfoYKMbdP+iKoRACMF4PnVaF219vZnilT14hHAHJvfhtg39C9ZWETZRsNY7TkcstVyC9T5P2zYuHM8xkTMRQ9GYn39xhZe7MYaqUOv4uEHM5ZbHW+208N8dYxmEGKXupnesW89DetwsR8sOqx2fIBIULJ1jx0dp+wEdL+b5xRodLyJrqBRMg+V2WkE3GNzV65ycKjDf9Gh5MX4U40YJYSxYanicX3U5WrIRSUKmqfHySpcgEliahqLE2KpG3tKIEuh5Ae0wdYtkDBUFgRevu6GAZgTeWpAGIiqQ0RSEqkKY0BWw0olodiN6MQO3k5KkP4aukbV0ekHIWif9nDUFwjhJXVoIKrZGtRsjAFVJCwSCSi+IEKrGeM5gPG9xdqVL04vRgCMlg7xtYPQCGm3oBOlxM1GAY5pECcw3enS9iDAWFCyNybyJ70d8+XwDUxFcqnvEiSBnakyWHCxdY3Y0w0rPx7F07CQha2m03Agr0bhjLMvbToxSypiESULe1llue8Ra2gLjTTMFHj+9yisr3fVKzya2qfO2u8e5azzLS8sdwiim2k7F55MX1ji/0qETJhQdk2rbxwtjzlQ7/O0ra0RhjBfGaKrKE+dqPHl+jUs1j7ytc8dYlr9zcpxHjlV4eLbEi0sdOn7Ii0upyK52fHpeRDlrpU0+hcpLS20mCml17s+8eJlqyydMBHePZrlnKse9U4VBzFja3HZjHSdJQtuLyNv6FW7ZYQFwNReNHyXM110WGi6gMF7wODa2sebumsynViIhuLSwOnjetWRH7XQTNWxR2c6VDdtbpq7HYi2R3Cik2DkgbGc12doUc6d+OxfWuizWXTQ17QY93Edn2H31yFwZU1exdI2J9W7SBcdIKxULd5PJfabiXHHB2uq2ylkaXpCw2HQx1AzgbyrFPyy4Ti93eOZiHSWO+db//GkqjsGpdsiaG1OwNCbyFl0/3eIF6fhqx2c8Zw3uWLd7/5ah8cixSpq9BBypONw9WaDlhjx5voYCXKq7oEAvXOKh2QqVbPqeL9R6XG56HK84vGGmRL0b4scqpiJoezGOpjDfdHl9rkgvSGj5qRDN2wamBndPZKn3Qi7UXPwgoS0EikjQNZUjJYeSo/HMfIvhJBcf0mAdIIljshY4toabxERxGoS8NXFOU6GSSwOna62AlU6YlohTQdcUWl5IFCYkIo3t6Xcy9yMIwhhFQMHWMAyLkpNWdj632sWNEqIYap2IhVZIO0hfOxYQhgnjRZ22H7PYdAnCNLvF0DXWeiFmKEiSEMdQabohUZzgBwqaBsWMyZlqm5yhoQKFbNpo07F0RnM2xyoObS/ipWobL0i4ZzK/nkgsiIGxvIOmacxUshQcjcm8hW3q3DGa5WLd5W9OrdINIsZyFpWMweW6hxcllC2du8azHCna+EHEUtNFVyBW4cRYgSBOe2XN17q0vZCuH3K07LDW9VFWFJbbHstNl0YvJEyalDNp/Njllsdk3iJj6WiqQtuLCWMXP4zw/IRSxuDeUoZH5sqEscJLS20mCxEnJ3KbRIsQYhA/dmI0xzvuHSNM2LZC8dVcNJaucqRks1C30FWFo6XNNwP9ysQAJ2fHCD773zE0FWWnbIVt2IuF+VosUzejD+CNQlqgDh9S7BwQtrsQbGdd2brw/DDmaxcbLLU9JvIWsxVnMGbzMRk0zbxrMs/c6EZp9pMTOSYKFl+fb2yY3CvO9taldXeVF8S03RDLSONUcrZGJWuhquGm8X1X1cVal/m6S5QkFE48xFuOV7jjQgOx2uWOkQxBGHNmpYuta6y0fRQ1xAuTwbx3Enpnqt20DUCU0PJCzq12matkWGp5rHYCMqbOsVGHxUaagXNhrcNMeZyJgsVSy2M0Y6BrKt//0DRxAouNHqerbcI4ZrEVkLF17hjJkghBvefjRxGjWQvL1DANk7vGbepuwoLbJRQquiI4WnGwDIXVXki8SwkeT4AaQc7SKNuw2o2viKcwgKKtogkwVQ3H0ijYGp0gIe9o6JpK6KduqTDZqIUUx9DsBqCmAsZSVQqWStONWO0GtP0ETVVouiGdQOAGfaGZ0vJhBhjPmXS9gMCPCYC1dkDLDcjaBtMFh9kRGzeIWev5NLohS+2ArKVx72SB4yMZdFVlqelRsA1KGYtiRiNjGbxUbVNt+tTdEMvQmCvbrHQDFhoe00WHbz8xyiurHVQgiCJWehGN3jK1bsxCw6ftBRiqSjFjYOgKSSKo5EymiqmIfeZinWonoGzrTFay2LpCJWdjaYJPf2MptfRlTAxNoeun3/leELHY8FhseYxkTEazAttQuXMki2OlLSZQYKnpkjU0LrsekDBbyfCeB48wXXL4xkKTlhsRxj1mKg6OqQ829JYbcma1w1onANFmNGfRDaPrtnIoKEyvxw+dGM/uuDkruo752DuuWDtX28z30lNvN8tUPxbwWo53EJAWqMPJdYmdOI755Cc/yV/91V9RrVZJthRo+NznPndDJvdaYrsLwV7uhPqbk4KSFuPbEiuz3cVl612joigUHGOzyd3QrriI9QOCTy93+OKZVY6NZNIgVVtnsuBwYjxL24s2CZ3//kKVMyttbE1FUwQLLZcvngmpdwMemC7y6LEypqHx5Pkad47lQIHxgo2CMuhOvWPg87r7brXt8YXTK7hRwstLHaI4oePH2KbGaM7koZkiT5yrsdhwMTWNU8st6j2f+TWXi80eRTt17+RsHUVTCRIFFJU4AU3VGC2YaSPH+Tp+KFjr+Nx3pIiCIG+b6Cp0g4g4TsiaOhlTp+nHuIFIi/KJBC8CW08LtEXJhmtLBRRV5Z4Jm7OrHcJE0PMFnTA1AKlAHAsWGj06QYSpq6iKQsZSyZs6WdvAj1zaXno8nfQrEIu0VoahgQasdj1eWFbWM+0iMrog56SxM2EckQjI6GnwdEgavNzxIo6UMqgKnBEdGr00ZiiJYNJUmMzr+GGacaUo4MUCPxIIBJdqPRLA1BVioZAxdSZLJiVToxekvdYW1kXDqeU2XT8kjAUjWYtT1Q5vOlqk5Qa8Uu3QDQS9IOLERI6jRYsjJQs31LlnKs/xisOL8y1afsj5NZezq11AwQ1jmr0QN9Cotn0UTWUq7+MYKgtNnyCOGSuoTJczA6EeJTBdsnEMFRQoOiaGpnK07DA3ksU2NJIk4cxyh6fPr/HspQZumKAoCrNlG03XGM9bLNRd2l7IhbXu4HmKopC3de4czRJFCTPlTOpq3cESejUh4kcJ1Y6Pv97F/YXL7W1dyNux1818rxlPO1mhLm3Tzf12yKC6nSxQkr1zXWLnn//zf84nP/lJ3v3ud/PAAw8c2C/t7cR2F4K93AnZhsYbZ8uMF1K31FYRs9eLy05jh49nagptL+JMtU0viLF1JQ0AXS9IeHq5s6mJ4mTe4sxqh9VuQNbUODaSwXN9TnzqjzE1lS/83/4ntAeOct90nqliBkVJqzj3YyXKGTMNgoZBF2nYSJu/sNaj2vLo+hGWodELE3pBwEo7jb/RVJW7JnIgBBN5EwR4UUy1nW4kNddHJe2qvtzy0uwZP6Li6IjEJLRjICGOIGepzDfSeJ+mFTJWsOj5Mc8vtFhqeQRxsu7C0pkbSdPSX1l1mSo42FqMQGWxmVoPgkQQxQpeECEE1DsetY6HroKlK0wVTZZbAd0gdWtVuwKFmJWOi66SZksZGnlb581zJZ65KGi7HdphKn619WLGWS0VV4oChmHghWkguEABVaEXCIIoRggw09ZJ2CqoCSgx1L2IasfjxGiGphvihT3C9RikxXpAz2+iKApNNyZRBFldQVdBoKIKQbuXWm1GsgY5W+Xlyy26foyuaRRshaJtoCiCIAhouiqmruKFIRfWAi6utnhlxaXW9QgShamiRZwI3nLnGD86lTbnNDSF//aNJZp+hEjSGh5RIrh7Is9K22Oh4RHGMaGA8azBubWQ5VbAudUulq4xmo/JmDpTxQwnxrOcjBIurPVYbnvkTGMgRlY6AcdGUsuJ64ecXe2w1PK5VHNxzDRd/psLHSIEpXXh3Fu3uC40PGbKmcFmP1fJIIC5ioOqqANXLeuxdbC3tPPha0PZMan1fNpezLabcxjCv/236eMPfAAfdc+b+fVmPO0kGG5FBtW1uqRuFwuU5Nq4LrHzH/7Df+BP/uRPeNe73nWj5/OaZjuLy9XEiqIoV7ildjvmtbz+VoJYkLd1TozlqHZ8RnIWNTek2g2o9wLG8hYLDQ8QjBdsZsoOd45mWW2noiJj6txZNPmhP/wYAH/wQ/+Qta5H23MGMUpJkvDlV2q4QbxeFDAZ3CH2awFV2z4lx6DeC8jbBqM5k/GcxanlFtVWwPOLLRzT4I2zRS43evzXry/QdENGshaPHC+jCIWOH6OQWquMOK1CO1/r8spKB1ODvKXRchN6XsRTF9dQhMBPxHradkLDC7nc9LhY9/DDGF2Fgpk6kda6Ia+fTjuWLzV8/BiOFlRqPY1aN6QXJpQyGoqAVRfEegyPBmRiQTGjYFsaYRyTxKmFR5BmWYUJmKpAKDHVVph2U9c0kv4YUhdW3koLBQZRKo6iMCJnaZAoGKaKG8T4cYiqqOiqgqoJuj6g9mN+oO1GnKl2CcIQY91omCipBcmLYamdFq1TtNR1JkyNyfU09IaXoBtp9txM2ebsapoKHUQJUyWbRJiM5jT8KKbeDXj+cpe5kokKXFhzafsBUSSI4wTT0uj5AattlS+fXaGQMbl/usDXLjb5yvkatq6iZCzeOFvi7943yX1TeZ692ODcWg8vipiyDTKmxlLLQ1WhaGuM5G3eerzCtxwrk3eMtAq4oXFiPEs5a1DJGLyyuv69W++Gvtz26PkRYRSz1HBRVUFCwmzFwYtCemG6sY4XbKptn44X4QZpAO9sxRlYY+IEVrshj8yVmR3JcLHW48kLdSby9raxctutySvrTakoirdJOA2uBUEAH/xg+vh//p+xMpk9beZ7SSnfiYMiGK7HJXW7WKAk18Z1iR3TNDlx4sSNnotkG/YiVvbrbmnrhc3S1YEF5tsck5mSxV88v8xi00dXFY4UHRq9tJ7N0XIar/DtJ0cRQBglZG2Dt9w/OTj+XCWzKSj6xHiW02vuerZYGmyNEINMs36VZy9MM6kqmTTDrOyY3DHq8NBcif/y7CIr7QBVVTB0jcVmm1PLHbwwwQsTZhoO90wXsXSVvK2RiARFCJ6br/NytUO1nRamu2PExo+h4YXY9dQ6UTA1RBylbp3lFnECXT+1nmhqGohb7yU8t9jiUr2HgcJi2ycREIYmvSDGT8DQBI6h0exFAzckpEKlHcHpFR9DTdPIR7IKbV8QJ+An6wHECbiBIIh8njlXQ6iC4XZXAetBxkEqSvREpO4sRZC1FJZbEb0wwdA0RrI6UZKw0g0xtNRqEwOanoqmnh9zdsWlnLOwdA1dE7STBH/dc60BpkjnlIQRS+0IU9eJYoGupX22zq+6nK22CeMEQ1OYK9k8eqzMU/NNVtoJXT/C1lWafkzLT4taumFM1jQxtfT7t9YNabod2n6EqmmEkeDFxTptN0TTFB4+WuKxeya4cyxLGAsaXsR00ebcSocgEcxXO9R7Abah8z0PHuGxe8dY7UV8Y7GFFyRYhrJexqDH6eUud0/keey+cWZH0m7oT56vcbnhstTyGC9YFLMWmqqg6zpvmCkyls8M3K7DVqJqx98klrwgIW9rA1exHyVU2/6GuBmKlRvPW4Nsyu3aMgwHNvdjZIaF027uqRPj2UEBz53i4XZKKV9quVQyFvdN5wetKbZ7jYMgGK7XJSVr+Bw+rkvs/MzP/Ay//du/ze/+7u9K1XsI2OkObrs7oq01Nt40Vxlkdp2cyHFyMo8Cg+wpx9S5YzQ3OE4xs/F9iQWsdX3yiUEYp66matsnZ+sYmspM2eZifUP8nBjLoqgK1Xa6gcyUbOKVhOcWm3xzocG9U3lmK1liAXdP5Hn0WJlaJ6Bg62hqzETBYmYkQ97WKWdMLjc8EpFma612fOrdiCRJXWRhInDDNAZlpetjmypzozkcXWGx4bLaDfFFQt5Km65ahkIQJrTcEFUFhYSKYxAkCcr6a+QsFUtLLSe1jo83VNbYII2VgXXrTJLG9hwfd3ido3Gq2mGxnUojXU3/FgtY7aStJ7a2FGoM9ctRYhCKQseL6YYxDS/B0mC6YHD3VInn5hv0wpBoPY4oo6bzUVXorQc+G15IKWOiK0DsEfkbr6kkoGvQDlJXUpjEVDIG41mTvG3wjfkafpC6k7K6ihtGfPalKi9d7oCSFi8sOjqqIggigaYJcqrBWN6gFwqSJCYWgjiCrp+Wk16spTVmVrseWV2n5Kj8+XPLvHi5xUjOotH1eHGhwULTJRYKigqgMFEwGc1bGIbBSqfHWifgctNFAXpBxIW1Hn4sqPcC7p3OcbScBRRcP+alpTZBnFAwNR46UmS+5VG0De4Yz3NsJIMgg9pPzTb1Qb0qIQRPXajT9mLytsaDR0uDispXWED6sXKVtM/XU1uEixCCU0tt5tfX3F2T+UE6uKIoA+E0XEZi6xW6H9i/a+LDDsUOl1ouLy91CJMWAjEo9LkdB0EwHBQLk+TWs2ex8/f//t/f9PvnPvc5/uIv/oL7778fY0tzjj/90z+9MbM7BBz0FMadRI0fJVxu9qi2fJIk2dHnvtWF5pjb1MIZvsPr9QZ/i5L0jrXeDeh4EYsNl6Ktc2mti6IonF3pUXcD8o6BratMFixMQ2O27HCx5vKlV2pcrKUByvVuyItLbTK6wljeZm4kQzFr8egdFVbaacHB46MZKtm08m8QpjWBspZO3jRwHZORbETDhaJtMJK1CRMfBQVd0dKMpUhwz2SeY6M5nrpQJ4pjLF3hwmqPjidoueG68EkL2Vm6QkaDTpikViAvRtXS4Gc/TkUQpOJnuz6FEXC57hJGFl6UxtUkAixTxUwEiUgtOjHrbqQdPmMVKJoKnSDCDRKEgF4ALS9ASdICicOFrXsJ5BUI1i1JCamFZq5sESWCrh+g+ckg3d8njfPJWemGq6oquqZS64Wcr3dZbocE64IobSfRo9EL6AUCU1cYyTrcP5Wn5UecrXbxgwRdV9E1jUlbYbmZusNGsiaTJZu1bsCffG2eC6tdghgq2ZhvLLQJopiLax2ypkLNjTi/2kURgqJjUM5ZFB2TmYrD0UoqeNOSCwJDU7hQ62HrGqqqoMdpxtV/fnqe+6fL/P/Z+/MgybK8vhf8nLtf391jj8iM3LPW7q6mqze6aQS0hMT0PCRkQkgwBiOZZCB72pCZNGAwmBDiPenNSJiebJCAeZikGT17jIQQeg9oSYil6Ybq6q22rKrcMzL2CN/97veeM38c99hyqcyq6u6q7PiZpWV4hLvHDXe/537P7/ddvuXCFJapgUmUFnSjnO94fJY/O1XCMuC1rZDnbrRJc/Ac4w63ZKXUoQtuzdfr5YSHdn62zGzVOSQgQGje2tGORJwVfGmly1pvbAo6VdqTmE8u7FKGDKOcr9zuslAvcaEi9gDPZD16s5LyVsklk4N7Gn0e/B3vhDXvndJhOq6vfz0w2KnX64du/5k/82fe9oN51OrdIGG8l6mgbcCN7ZBXtgY8NVfj4/eYWh4FPzoQsTjU3Tl4n4N29VXPZKZeoRumxJnkK6t9miWbXELDN/f4EgBRWvAbL2+SS3jmRJ1ulHJ9Z8TGIMYWgkJJdroRw1RyqlVweqbC6ekKCzWPp5fqRFlBnEkKBWku6UYpnq25IEsND4TCtUyUgCnfYnuUopT2sTk55dELMqqOiW8Lnj3ZYBSlvLrZ5/puzGjc5bAB14JW2WWp4XF1J2KQSAylQUpUQMlQ5EpQSN2hUVKDmglQsdHPNdE3bkeKbhQjGKukAF8pLMtgFBZ73SCDe4MdISApcrJMUWS6I1QAK31J99Uu6V2QVlDsP58B5CmsdQOGmWAQ6awtk30FWNmGEw0P17bIcoVpCTaGKYMwIyk0AKx7NpYJ3TAlSHVHzDQglwXtICPO9FhPIjCFIEwz+lFBN8hwLDg7XebxuQr/9fVdLm+NdK6XBY5wiJOcXpQzSjJqnkk31CaIhVI0EJQdk8fHncflpk9aKM0Ta/kURcFrmwHDJOXMdInLWyNeWu9xqxPh2BYfPN1gpuqxUPdJCkXFs+knOY5t8juv7fC7V3aouRatss1SU9skHAUQyy3tXTX53p7/VdVFSXWI2H9xrnpvJeX+mXbk9v54ahhnvLI+wDb1+XfSdZm461zdHnHu9P05O/eLnnlioUKSF4yS7K5Gn/db894qCHqzj38ndJiO6+tfDwx2fvmXf/mreRyPZL0bJIyTRfWoqeBs1SHIc0qOSZDnjJKCeun+xz5psX/xVodcKj50psVj87W9hUlKyaX1Pk+P798oObTTHKUEvmMySjRXwxSK3VGKnxScaPq8d6nOF252eHltwCjNUUry1GKdXMF81aUfpqC00qjpWbiOwXTZ5lY74Eu3umwPE1olB8sU5IWiF2W0Si4VN6ITKLppzEzFpeqZDOKctFAESY5Qkrl6mU9enOKLKwO+strny7dy/uham81RqgMmDwIFoaXWtlDcbEes9ZJD3jcKCBMQQmIb+rZU+gcT0JLBHWOHyfcMNMDICkVaFMQH7nPUzqdiar5Ojs7CChJFkGpvn733Cxhld18EJiAmH38dKljr5yT5/rFOfqcAGp6+oPiOQWJolZeFQgiFIbTPz1zdZbsfMYo14HVNqNoGUaq43R3hmhYGippn4FgmUZLTCzJ2I4ljwPXtIXMVi9VuQDp+YT3fZpTmJFIRJgWFVARZwVLDpx/n+nUXsBNkvL41xLZMUFAem2lKJXn+Zg/LgKeW6jRLDpe3AxqeCwYs1jw2BzG9KOWZk3WeWqgxSjOavsMwSrnZHpFmObtZwZMLlT3DSsfU5HfHFFzZHrHajTjR9Lk4Vz3keJ7mkjTXwZ2moW0XJiGlBwHSZIQF6DDbmseJpr8XDzGptFAEaY5tGmRS0iq5hz5P28OYk4U6RG6+XyTMQbAipeTSxlDHbJQ1N+kokFFK3dM9+fLW8NDr8DCA5d2wcTyud3Ydmwp+FevdMC++l6ngyabHhRm9IJ2frlD13vijkuSS292Ql9YGBKm+RJ+aKuM7FkopLq0P+dy19h7Y2RkmhLYmCr93scZzN7pc2x1hCRgmOZlns9qNONUqMVfz8RwThcI2DZ6Yr2AIg81ByNYwpeTZLNkWJxueTvEuFOs7oz07/Zmay2LdY2eU0vBtzk773GwHGAK2hyntIKPsGPiOyVaouzoIg2Gc88XbfS5tjtgJMjqh5uRkWUF+AOiY6ADOMFXc6qQIY59QbLDfeSkkWJZWSZUcqJX0SCtIIRqjh7t1aCaAKUcTgu9XFjr0s+RAJxq/N4UeIx2dlyk0ofluv+9oje42axs/fjNQDPOQNAfXlLiWRdkVlBwDT0LJtRBS0QuL/UMoIMoVjpAYWAzSDNcwqNgG83WPm92IRI3/bgm3+zHi5gDHFDgGNMsWjZJFJ8yIcomFwLFACINTUx6ztRKdIOXydoBjwiApeHF1wOvbIz56dkr7NHUivrLapeLYNH2bTEpsQ3Cy6bHQLFF2Tb5yu49rCVY7kmbJ1tL7Xg8ATxO0UFLhOxbPnmri2SZXtkas9iKmSzZfWe2x3o/ZHuiA24M8tHMzOng3LSRVy2Sm4mAbd8rPD17sZyoOH1je5/0cLNfSYzSAZsnhycXqodHxzJj0PLnvUdf2YZyz0Q/phjlKqUNg5dL6kM9c3cU2BI/NC9KigmdMNjFDOmHCXE1njQGH3ZOzgi/d6rLW03/7qVYJz3nwy8+7YeN4XO/selNg5/3vf/89Wf6e53H+/Hl+6Id+iG/7tm97ywf4bq63Y178tZh/381U0HMsPvnkLB+OW3s5Pm90fBNw5zsmArDNfa5AkkvaQYy0Hf73n/2XPLVQZ3GuwXZU6EXRsfAc7Yh7eWNAmBXk3QjPNLhSdXlyoQpiia2BTmtf7ScMk3Qcg9HgRidiueHjWoIb7ZDfeHmTkqPbJ4sNj9OtEkmmOSGbvZC1TkgnSEApTEPnWY1inejtWgYzFYfdIGM3iPnySs4g1pLhAjhddcnKDoUc7Zn5mWMjPzVRKSlwTC0rn/BexPj/vNDdlmECBZLzU2XKrsGrW0M64T45WKEB0GRJn4CE0VE28pHKgW7CHqdGMDYxvIub871wk+LuPKJ7VVzog5UKIkCRU46h6psoIYjSgh0liQ+oxnKgH0vsTGIJicBkM0xwbYNenFEvOUyXLJI8pcghyxX9KMG3TNyGxcmWxzApiHoRQph4rkBiYAlIC4P3LNVplBy+vNJFKnhte8goykmjgps7I3y7jmOCZ5kIoWNXXt0cMEwKzk6VyfKCODfZGsQUUtIZZbiOJqJbpoFrmzwxX+VEXCLMCm52IrJCIYTkSytdVrsRSklWuiGjuMA1BYMoZWsQU/VtfNtkse6zPUx4bK5ClOas9aI9Q82Dvjmg86b6UcZqJxxbO5Tu6HDcdc3xPNR/+k9kuUTZ7h7peXksc+9HGUmmc+E6QcKN7ZAwzynZVRxz3429EybYhu4YNUvOXgTGK+uDMQjSa8SHTrfusMLY/5wdvf1g9W7YOB7XO7veFNj5k3/yT/LzP//zvOc97+FDH/oQAM8//zwvvvgiP/RDP8SlS5f45Cc/ya/+6q/y3d/93W/rAb/b6q3Mi99M6/atzLWPLpITEJTkEleoO57vUPZW1dW+OjMlvvuZRXaHCUtNf+9+jilIMoU0TcR3fYrlp+cQQjA/DkVUSjGKC3YGsR5H+TZRoRglOX9wrY0hDJ5arHFhrgpK8fmbHS5vacfhj5+b4iNnWihgtRfz6uaIKM0ZxTBXdTg9XSLOcv7fn73Jjd2AuXGXx7NtZis2V7bhdjegZFucbPqcXKjyyvqA3WFEmEkSt2C55dEJMoTQPiqmabLZC1FIzLFMPM8hZr+zcrphc6ObkRxBDYYaS8iBUQIrvZj3nKjT8B2SLCWV4Fh6FJVJPWoK3wDg3K0m2EbBnlT8q1lC6b87yvUlLc6hIQRBnhPGevx1kAc06VRRwE4gMQ3NqZKpPljPlvi2RdMvGETa00cWklSAZZqMEkmWjUdrqsAyLVolQZhLbnUCekFKVkg6YcpC1eWpuQqvbI1IM52R1QsTyo6NY2ny8e1uSD/K2RkkmAIs0+A9ZYdmycYxBSvdiPYo011HCVXX4r2LVd6zVOd6O+T8dIWKazKKdVekkFpWbxtacZZLxUtrA6QS1Mau4xXPIskUGz3t6bM9SrS7+FjlNJGfTy72aS4ZhCn9MEUptRdJcTebiL3bloX41KeQWcH2zc5ed+Rky2e26rLaCekUiu1BjO9ahLnOHfMcg7RQeMbhjtFEdj5xVe+GqQ5JHY/N7pZjdz/z0wf6bB0TjY/rLdabAju7u7v8nb/zd/jJn/zJQ9//mZ/5GW7dusV//s//mZ/6qZ/iH/yDf/AND3Yepo4uWA8UF3GAEOxYxhtKSu9XR4HZwfb0QYXJpCbH148yVtshz11vY5sG719u8M3npljpRnueH8stH88xWKj7VH09ZlrphHtmgen4AuWYBk/MV2lVXIpCsdINsYRgcxBxdqaEYWin3WbJIZNK/6wfcXNnxJXtEednynzLuSa/9WrO9Z0RWVHwuattyo7BjZ2AtX7MMEoppKLqaWpwmhekuSLLUwqldDJ8LyJKJJkE0xc8Plfh6vaIm7shX1jpMlPxyMe+P64B5jifyZa6o+MCm4OM8C7tkYO4QwOegkvrA/0NQyCkIkxgdPB9fpD3D3AYh40+ZFnsmxc+SDUtbTDYP9CpiQC3OMDlUdAOcqL88N9sAL6ljQuV1AotlO4+KQWua2AaBi3fJshyikL7DKUSBnGBQYFhFXTDnLqrH2+I8WNtk36SsTWI+c2X10lyxc1OhGEIPnCqzjMn6vze5Q6vbg7ZGiacnymTZgrPEhRKIcbkqE6Y4nQDTjW98XmpWG743O7BxiDBMnU8i2GafOLCNN90Ssvtr+2EbA4ipiou0xWH252Qm+2ImapgpuIySHKmys6e/DzJJY4FUip2hymFFJxsCC7M6nPtoPz87LRPzTOIk5Q/utWj5dt7/JeD5/352fJd1wHHFNQ9i1GcMVNx8GyT5VaJ1W5EmOSMkpyqa1GeqeLa2s180tm5F9i469jsTZifPkgdE42P663UmwI7v/Irv8IXv/jFO77/fd/3fXzgAx/gF3/xF/kLf+Ev8E/+yT95ywf4qNcE4GgX1MML1Bu1bieE4C+tdAF4erF+R/v7rXSVjranjz7fwd1mR6ZsDRKE0LyAVtlhaxDvHcvJls98zcfIc87/5q9ilGxWP/ydhBikuURJiUCQSsl7TjS4MFfFNuCF2wNe3xqwM0j4zJVdfMdkvubrsZaCl9Z6XN8Z8QdXdtga6oiJ73lmCc8SeJbBSidiFGcs1D18x6DsGAhgZxBr4qzp6iBJpSg7gqvbI7aHCWuDWLsVmzBTcXl9J+S1jRGDTF+sh5HmQSgACQ5KX7TRYyqA4AAQuF/FBVhJzlTJxDUN2mFB9CY6ORMHZZM7fXfeqB5mZAU6O8sUAitThx6bHEBLORziNU1KAuFYSn+iboOh2BzkmGhV12LLJ4xzemFGLnSuVpDqjleQgS1AFBJhShzTYrpkkOSKimtiCQOlBHEuudkOSPKCdqh5ONe3Q85OVWiUbFSIJlGbuiMRZgVN3+HifJXLm0Nut2PWexH/8cUNTrbKPLVY41svTvMHV9sEcc7uKNXmgBWbq9sBvViT3iexDVXP5LG5ig6jdU0MFLujhNu9mA+faVFxtYxcSZ3jtj5MqPs2Sw2Xp5fqeI6157jcjzLiNOfy1oBrOyM2+iG9ICfNC1a7EQv1w67LczX38CapZuP+b/8rW/2I3734CXZTDYwvzlXHDtcltoYxF+eqnJoqYZuCVzdGdMKEK1sjTk2V9ro1d0tnP0h2Tgt11y7w5L7HYOW4vl71psCO53l87nOfu8NF+XOf+xye5wG6KzD5+rjuXgfHQPfKt7lf6zbJJau9aC+iYabqjnk39w/QfNCgwTva00eeb2+ha/nc3A20fFZBVhQ8f6Ot2/W+9jPxxlb8s0ZG42/9CAAb/+Ulcq/E+07W2RjEbA9jSo6JbYg98uTmIGIQZ8xWXK7tBizW/T1p77mZEp+/0eb1jQHXdgOiVMvLX1ztMTe+X8Mz6QQpaaF4cr7K6VaJ17eGmkMyTOg3fCq2xWzVxRaKrFBc2hjQHqUIoX1TFhouL652GY3Bi/ad0V+baF+aNDl8oU8eAm0oYJhB3C9wDQjewsgpR3N+4O6Axxz/vrc61Qoz7QTddGHnSCvJYRx4qu4NohSal1SpeNRsE0VAe5RRKOgEKXGcERfguSY1z8Y2IRh3iORYyo8EQ+R8/GyDUaLYGMRsDbRRpACiTOFbmiRtCDjR9Lg4V8GzTJ5f6eKaJnXX5j2LFt1RwpV2yDDJOdPSnZDVDoR5oV2bheDJxTrCMDENTa6XEn7rlW3CrGCx7vHkYo2piocQOpCz5tl74+AgLfBtE8cyMAVcWh/y8nqfrJAYQrHc9JFK8fELszy5WN3bAE3GTFtpwWo3JC8kN3ZCLEOT6Wer7p5n0GRTVPWsvW7pbMXFlTniL/0lFoCV/+UzDC2Py1sjPnoup15y7hr4241SBlHOWrfLai+6Kz9oUpOx2cOM3N9JXjzH9Y1Rbwrs/PW//tf54R/+Yb74xS/ywQ9+ENCcnV/6pV/ix3/8xwH49Kc/zTPPPPO2HeijWIc9bhTNkrO3UN4rofxguZbBiYa/p+w42SpxYbaid1f3WEQelAf0MO1pz7F4bKHGqekycZLxv3z2Bi+sDqh5Fn/xQyc5N6MJlle2Rrzw+hbfO37saxtDemZCJiWWYeiWuYJb3Yj5uscXb3X4zOVd+knK+080eGrMY5ip6Pb6y5shtzrazbju2ViGNhSsl21KrslS0yP0DYabWuIrDIO//PET/MvP3OQrK31GUcbLqz2iXGGhKNkG20FEL8yJc32hTsfkWds0EWOIYDJOI2e/myIflnF5l8rYB1FHa0I2fpA66NNzt5+9HRVJTXjO73KdSgFL3f93eQKWmy6Pz1aIs5x2YGAY2pBwEOckqVZgGabCNQ3GQeR3ALUwhX6sUAL6UUGS5hgGeJ7FVMmh6ducNLQ3UJwpru+GnGi4LLV8PNNguuqSZJJPX9pid5SSTvl8YLnO+08t8cp6n9c2hzi2yWzNxXMs3r/cAAW/e3mLtV7E7W7IKC3Y7McsT5X5lgtVMsne+ftNp5rMVF3yQtIeJSSFYrrisNoLWe1GSCWRUjGKcxplB98SXN0J2B4mOiur6bPajQhinRS/M0wouxatks17lhqcn61gGMad8Q9Cn+tpLlFqXyhweqpMRzg8NlfdU1geXWMOdmyHcUaUFWwODjgy32UdeBi11FdLRn4MoI7rfvWmwM5P/MRPcObMGf75P//n/Jt/828AeOyxx/jFX/xF/uJf/IsA/PAP/zA/8iM/8vYd6SNYR8dU52fL9wUqR2syB1+eKh0y8fPuI1R40EXpfoTAg6O3g8frOxZxmnOzE2n3WiH49RfWQZg8c7LGai9io7/vDtNPckaGwc4wYalZGitTYjZ7IbfaDoM4oxulOrkbRZzqThZK0fBMtvsRS00PqQqemK+gEMzWXGYrLjc6EVmuSHMt946LgrxQtIOCsmNRL9ukaUFcKIIkYxjn+LYehxQoTTiWWm79u1e6nJ8qE5UVCqVVLYXmd2hy7MOrSx62HqYTMzH7+2pXAvf8w+83FvOAp5ZKfNtjc2SF4r+81md7mI25OQVl20CaGsCUbIPllkc/SrGifE/KPzFi9E0YJRm+beM5JlGcE+QQpDmGUlyYLWMZJpKYIMl4/kabG1WPTCo822BrkKCkJM3lXujs52/1+IRr8/Rig16UaesAwZ5bcMk1kGNzSiWg4lg0SzZCaUAzXdUd7SSXXJitcGqqjCUUv/XyFtd2hnzxZhfTFJhCoaRirRfRj3IKqbjVjXAsc5z/FjFXdfSGZmTwHVNl1nohK52QOJc8tVTfk4UfHIEvt3y2BjE3dgMub48wFks8NX7t//LHz5C4PjXffsNcq0nW1tYgPuTBNQEnUkqGY4HBw6il3qyM/H5g5u0EUMeg6dGsN+2z8/3f//18//d//z1/7vv+PX92XLruBijuB1Tu9Rz+Q/hVPAgP6OCJfnQRmpCWd0cRo6Sg6lnakv6APf5y0+f6zogkL1hph/y7L67QCeY4N1tmt+buPdfplo+oVJireTxzsk5nlPDa1oB2kNIJUpqexcm6j2XB6ekyG4OIbpjy6UtDPnOljWHA4/MVPnxmGsfUi6gCRnGObcBuoM0GfUtQ8xxKjsHztzrYlsETs1UyKdnsxwziAsc0yfIC0zCxTYNa1eBmOyVHq6FyVfCh000sw2B7GLE9TLQyJs7IxmOcKQ+68d19a76W9dUGXg9aFvvS970cLbSjczfIuN0Z0Q4KNnoxo1iSoUdwColr6QgKzzKQEpSSOCbIAiwbkkzL/X1bh8wKUVCyDQJHEMSKAtgZFtzuxFxcqDFX9bjdixkkEiUS6p7NRi8hzXUmW9W3ibICS2jN/6WNIY/PKcJUcqsbkkqYq3kst0qs9zRgX2r4vLdco5CKQZzxuWttnrvZ4yNnW5xolugECVXP4X0nanTCjJVuSDvMeGVzyKlmiYW6R7NksTNKGcUFtmUwU3ap+Bbbw0QDjLU+czWPZ081cUzd+fIdkyCRBKlOpT+alH6y5e8R+G3DoBvuzxl918Yvu7xRTdaVi3NV5mseX7ndpT3Sn+wJP+e/Xtrm6u6I89MVPvnk7AOrpR4UGB1ci+BO76G7iSXeqg/PsXnho1vHpoJf53oQ0t7dIhjeyu+726I0+R2TndzdUo33SMtXdtgd6YXl1FQZ2Ccve47Fp963SLM0btV3IhSClW7Ixy9MM//k3N7zzdU8/uSHljENg1vtkE+/tsPV7QDX1iRi24BIFlSURcmxODftEOcDtgYxQRLj2QYl18azBFvDhKvbI2xTXwxavs1S3aM9ionzgiiXDJOMQqk9o8L5qsv7T9b48uqA9ighzRSpVNQ9k6fnyvz/vrLJ1jDTnJpxiOaHl6sst3z+6GaXUZzuOeAqAY2KA6RsHbQ2/jrUw5KNv1rlm2MPIjSfKR/7FEUKrnczdoMdPEsQJOqQg7RKQdnaq2hrkNIOuvTTA04t2RjQKYhjKGRMreLQLDksmj7Jbqjl70Kn1s9XPU5N+ax2Ym73QoKkIEhztgchnVFC069zbqqEYxpESc7mIEIB56dKVFyThufQDhL+/ZfWODdTZq7msdTwyXJJo2SxPUhoBxnXdwIk0B4lnJ0us9GPUSieu1rh3FwZ3zSZKtnIAuxxYrllmFQ8m9mqy8W5GkGWUy3ZvHepzotr/bHJZ8KpVolrOyGdUAMoITJGiWS9F+DZMF3WTK0JeDg/UwaFTmK39vt8SimScSbXg3aPqxN5/CCm7Fg4pmAY51zdHY0B0IgPxy3qJecNc7IORlGcbPkwXne8I+vaUdBxFNDdSyzxVn14js0LH916YLDTarW4fPky09PTNJvN+54onU7nbTm4R7WO7ljuthhMbkspeWVtwEtrPYQQe9k5bxXwHM2zurI14nY3ZKsfEaTFoVTjyTEqpdOgBYJBlJNLya1OyNnpCraxH2z42HyNvIAT3RHXywGplJybrlDzbdJ0/yNnmybrg5TtYcJqJyBLC8qOSdm1qfo2G/2EYZgTWgWXtwN+8CPLTFVs8rxgtRdzZqrEY7NlPn+jx0trXXZHGVXf4kyrzHzdo1ayaY8SokxRFAWdUYoqCcIk50Y70KCu6vHtj81xfrbMle0htzoxuVR84vwUjm3x6y9v0BkljFL9PnRHKfM17X0SxAVRViCVvqD3wpTg7SLFPAIVy3FwqdCjvqOjtUEKUar2ANEkX8w0oOHbtIOMtICiONyrUke+7qQw7KT0Ak2on6lYxLmiZMH5mTKmkFzaHOKaBo/NVxmEKc/f6JBkBcIR9KIc1zDJpcIQEpTJKCnoRTnffG6aVzeGXN0JSAodVrtQ97g4W+WFtR5fWOmRFRJZSCxDYJkGlilQSruJWwZc3w253inz/pMN/vx7T9INM9Z7MTXXZHuU8PRilbKrOWfay0mw3NTdGUj3Aktvd0OirEApRcWxSNKYz17p8utfWeOphTrf801L1ErO3khrturywbkWXrqPvq9uj9gsoofqWqSFwrW1dN619ei66lmcn64Ao3s6rB8FLAcl8bMVF4Xiyys9AL5pubmX4A53AR0t/75g5u3y4bkXaDoebb3764HBzj/9p/+UarW69/XxG/7m6pARX8UFwR4Z8ehisNzyubI94ncv77DRi2lVnL3snLdztzFZWMJxNlWY5pQcS6caZwUrnWhv8ZyreqhFxVzN5VZ7NI4FMLi0MaQXZXu7sG6UstJN2A5TFuo+y1NaHXUwy+fJxRqdMCXKCh0BsVjnwnwVyxA0SjbdIKUTZmyP9HHYpuDKdkA7TCk7Jh8/32Kq4hJmOb5jY1vaw2SqbFPzbSzDoFVxKTkm/VghDKFzjEYRg7hgtmJTdQQ3d7Xqa2eUsVR3OTdTAaVwHIvH52p8Je2wNdT9kiDR93UNg0GaEacgDbCE5gaFDyg3n5BtH+WaSPFDpXk2k8XmYOfp4MtVsWG2YlH3HaJUsjPMHopQHacQoYiznOmKQdVzeel2jy/f6urO4FyVp+fK/P61HrujBFMYnJmqkBcwiFI2+zElEySSimuS5pJzMyWeHJtM/u7rO4S5JM0VnTAZj5oKPMfi8bkqHz1n49paffXZq7uEcYZEE/zjcRCtZ5lcnPPIcsV/fXWTlW7IXM3jU0/PcbOTsNIJ8GzB5S19PrXKLicaLrfaIduDGKUUlnBY7YS8vjXk89fb5IViox8zU/c4O13dU3UqpZivupAVe0Gg28OYgXC5W9fiXhf0iSHozkiP/ybjtDdyWD8KWA5K4tNckhVyL85lthZxanp/XbsDdLyBMhXeHmn73UDT8Wjr0agHBjs/+IM/yGAwIEkSvud7vuereUyPdB1cANKxKV2cSQ4uBv1QW8LfaAd0RgmuKfAck/nqOPzvLUjK73Zf1zKYrWiJq2+ZWr7qa4dXBQcWLPjgqSbLUyVu7o4YJTlK6ZTvbnjA36fl0yq5xHmPJJMM4pz1fsKZGYnruqT/9n9FAU+emeFqJ0GImAuzFU42fQSw0onYHER6dz9KmK64lByDjWFCkuVs9xOkgt+93OZ9J3JsIUBJpssOngmvbAy40Q74wKkmf+ziFA3f4upOwFLN5ZWtId0gox9K2sOcjX7KTG2Ebeh4i6tllxdWu6x1E9JCkucSKaHsmoRJQZbDzqigbBUYps65khNzvOLBicGPOtA5WJNcr0nn5p73UTCIc3phjhwbBZrjF8oV4Du66zOI9t2YD6au+45gN1CMchj1JCbRHrByopSk6LPdDbnWjigUnG55fPRsg41ByvXdkEIW3B7EpBlsjRJut4esdgOeOVlnd5TSDTMWGy4gCRPJajdAFfDB5Qanpsv045yFeompssmnL21TK+luzbOnm0SZoh2m/MbLmzy9VGetH3GjHXK7E7HRiymZgm5cEKYFaVZweWtEybaoeDFrvYhukFJxLWzLAKW43YvZHkSkhT6PMwn5ONKh6Tukud68/B8vbSDynG/+f/0yJ1o+M1N1ZJQfCit9I05MWihtwVDzDrkqG4ZBveTc830/ClgOSuQnnZ1uoHlAJxqH17V7dWreaEz2dtRR0HQ82no06qE4O41G44E+UEVx3Mu/Vx1cAI52diaLwUTymRUSIQTnZqt89JzLhdkKnmO9KUn5G5H9Tk2VWO1FRFlB1TV5YqHGdMVFCHHHDivJJTujlCSTRFkBSvNvIKbh2ziWwZOLVaSSvLzWRwjBUsNDSsnlbsT2s9+hf69tc2HOvmNBOz9rcHV7xPXtAaOkoOWbhInk0y9vYgtBxbMYRAVrvZCyY7HRj+iEGZ4FL62H7IxifFNnGp2brVBxLJ6Y0/yjk/WMly0TiSYzD1OJGqRYpmKqbNPeGTKKtc1/ybFYbLhUSy6zlkEwzi4KUh2F0HIEZc8iTjJytS9J/1oood4tZTFOaufeoaOMfzbM9b+Dj51I/FMFZVNwslnidi+iE8i9ro/vwJlpn6mKw/MrQ/qxfgcOBdIriJOca0FOIjW5ueZZVD2HimczjDNe21AMEz1+6w8ytgYZ19urfObqLlVPv7trXYvbnQgBrHciulFKN0iYqnr4rsn7FutcmCuRpgVBWnBmusxS3eP1rYi1QUg3SJmuuEyXHRbrHte2R0gp+Mpan2bJZZQWOCYsumWirMC1TbJC/z0l12Kp7pEWks4oxUThWiYIwXLLp15ymKt5KKnIpWJrELM1TOiMUjbOfoRPPjnH04t1lgt1yMRUd2xdNgfRXQ1JJzYUk7XgQfkwdwMsB2+DlsIruIOzM3n8/UDF16rjcpzL9WjUQ4Gd3/md39n7WinFd33Xd/FLv/RLLC0tve0H9qjW3U74U1PlQ4vBRPK5PUy4OFdluVXaWwwmROKjO5n77T4ehOznjp1UNwcRcSp5dWPAVNnjycXqHQuWO4lrUDrpuRdnPDtXIRu3969uB1yYq+w5ISsp9Tju9W12hgmtcSrycsuH8fMd5CvFac7N9ogvr/bphimPL9RwDYNkTB5+z1KdTphhCYgLLRuOsoJuUNAJU5K0IFQShSDMcrJC8JEzTS7MVTk7PcdmP+GFtT6DKB13WAosBEGUMEgkYQrCAEto7yNTwDCRPD5XYbbq8uJaHyUVw0RRVxlpqo0F7wbxHzaG4VGrnDdPmD74OAkMQ8U1GZDkh3+WpDqWohdp7szdShn7Ro+OCRXPolF2+Mpqn90gpRemWKZxKO5dolVfozglSg0sw2QUZygliLKCXpLRHWWMkoJulFH1HZRUvLjWJ8gyqp5FZ5TymWsdpIKSZeLaBqMkw7EE3/HYLFJqbk+z7JDliryQVBybJxYqTFd8bEuwM0q5OFflZNPnVjtgvRtyYa7Mf/e+ea7thmz0Y05NlTgzXUEBX7jV1RslBC3fZr0XszNMeHmtz4W5qraIyIpDwaIzVRcp0a7QRy7oR12SH6aTchSwHL39IMnn9+refK06Lse5XI9GPRTY+dZv/dZDt03T5CMf+Qhnz559Ww/qUa87Tvgji8FE8nkQBMH9dzL32308CNlvckLP1Vy+crvL5a2ATA73SMpHj/GpxRoCQTuItRki0AkTOuOMhL2FMSu41Q74tRfW2ejGOCLne25+gbNzVW79qf8z23FxB19ppuKwWC/xubyDUoJBmHJyqkzJtnhsvsrJhsdKL2a56ZFJeGWtz2o34FY7QACBbdIq2QgBG/0YKTURMs4k0xWHTz45yycuTnFrN+TFtT7r/ZggzuhFkjjXwKTqCC7OVZitetzqhIRJyivrCbNVm4ZncaunVVpBdP/3enJRPsjRmXSAHpDec1zjygCV6C7PwUqBKMvxLB23kWWSo4K4XOqQUqWgUTJZbnp4lsnrW0PiXI8rs0JiHHhuE1iouyxPlRhECRIxzpBKqXo2Qpm0B5Is0yBlruYQZQX9KKUTaN+cCZXl6cU6Ty7VSHNY74e0RxnLLY/H5nU0hGsaXNkZUXYsXEd3VhslF8cyOD2tvaziNOe3XtnixdUeDd/mhz56Gsc2YOzSPOHDzVQcbuyMMA1438kGtpLUPv1/MHPZhff8X4mFzsna6yJHGXEmqfnWXmbXZF056KXjmIJX1gd0w/SQ385Xk7z7Zte8t7uOoy7e/XUsPX+H1t1OriSXbA4iOkGGUurQTuZ+u68HJfsJoVPOp8oemRxiG0KTlMdmagcXNcPQo6pL6xrk3NyFQZSx1gsp2VUsofivl3Z4bWtAkuYMo5S1XsSskfDH/u9/HYA/+PJ1Bsb4uGq6jd4e6TTnP/HkDArJi2t9Gr7Dn3hylsVGiaqn+Te2aWCaJo8vVLg4W6EXJvznV7bZHEYUhXZ3GUQpvTAjSHPW+zFhVqCApudwZsbHFAanZyq0o4xemO51DAzANxRN36LhC27IglFakGQFmczZHuYP3ak5eP+Ct8/J+BupCsbhoXf5mW0IDAOEkHcFkRI9epSAHRdjE8Gchm+NfX4SVCH3Q0yB5brJH39qhqmSze9e6RJkObNVjyTLxxEvNrNVB4TBmekKHz/X4sZuhGeZ2KY2w4wzyVLT4489NsP5uSovrvXZGZmsdAN2hhFhLpkta3VfzbdZ78ZkheQrt/uUXGvP2wf0+d8JEvphjpKKm52AfpSzPdRJ6aenyri2SZoremGGb5sEac6sK/hT/+OPAvD6n//TbEt7b4Ox3PI1+XmkR+kToKOUIk5zfv/yLtfaAeemyiw0PP7gWhvb0OvFZP3YE1xU3UNd6Lej7te9+Wp3XI4VWI9WvePBztraGn/v7/09fvM3f5MwDDl//jy//Mu/zLPPPgvoD+RP/dRP8Yu/+Iv0ej0+9rGP8fM///NcuHDh63zkb385piBOJev9iJJt7iUST+p+GTUPQvabPMeTi1VNHhzv4FzLuOsOKy302Ko9SrmxE9CLUmzDwLUF7SDlta0B3SBFKkXdtym7Br6xT1FNUkW1ofN9bAOipGBjoPOxhGHw3c8s8e2Pz+FaOul8lBQ6HHGYEKU6BHG56XO7G7PRD2mVHabKFsM453Y3ZK7qMky1aV2SF7SHCUrA7jBmlCRYhkHJMak4Fn3LIHcKRK7JxkEOz93oIoQkycGwLLJMMowUR5c9m+Muzdeq7sWH6geSelnHRxwFkhbQ8qETaaAUZHC7l5AWiqKQBFlBN8iZXM/K2tmQasnlyk7IFSXoRBklUyCU4spOQJzBhh3x3qU6Zd/m3FSZfiRJCsVU1eFbzk/x+taIYZpzslFilOasdjUguLUbUozBSpgUqKJguVVBKe3qXLINru8GLNQ9Vtshr28OON0q88RChdOtErfbIXM1l+WGz6vJiFwqlFJIpRhEGd0wwXdMMqmoeg6mt3+er3Vj+obc3yw5lnZhz0p7n+vJuX59d8RzNzuAHp1LJbENg0xKmiUH1zL2NmDtUcJqR0dg3C9H62HrQaTgX42Oy7EC69Grtwx2vpofgG63y8c+9jG+7du+jd/8zd9kZmaGK1eu0Gw29+7zj//xP+af/bN/xr/6V/+KM2fO8JM/+ZN853d+J5cuXXrkgkjvpYo4WPfaCT1MG1YIwfnZyiETwzjNud0JGMQZUkrmaq5Obk4156AoFGXXIldKpz8HKVIqEIJnTzZwTUHdc5DBcO/3uI7g6cUaK92Iz11rszmImas6tEcJf3i9s7doSin5rZe3uNUJODdT0QCnHdAJFJe3bLaHCTd2NZdhEOesdwMQgrOzFb7j4gwvrvW53Y3ohRm7gxjLEtxqR3iOScO3aXgGoJBSO/IaQpAVil4qUYVWAMksR0o92jAt8MaJmjVHA6HOKP+6Oyd/I1egQIValXRwZGgBDZdD50kO5FIyiHKiXEdURLl+nx0D6p5NnOfsDhMsQ2gPHCTrg5w0zRiM32hDQskxWKj7mJbBlW09EnNMQdQqAYqpkk2U5by6MUKJEe8/0WC+7tIPXV5c6aOEYqHms9T02AlSqp4OUD1V9ynbgudutNnsxTTKLt//wZO4pkGz4uBYJoYhaPo2G/2IimNyq60ztQZRxqmWz3zd56nFGteKfQdlKRUbwwhnnDc2qdtje4m9uImh7jDVPG1BIZRCILg4V97j8onxOCxOdcyGVIqKZ7E11OuObUA7SJkqOxiG8aa6JHotOpz79SBA5K12ZY4VWI9ePRTYOSo5j+OYH/7hH6ZcLh/6/q/+6q++9SMD/tE/+kecPHmSX/7lX9773pkzZ/a+Vkrxcz/3c/zET/wE3/3d3w3Av/7X/5q5uTl+7dd+je/7vu97W47jnVIPoop4M3PsN1JqKaW42Q54aa1PP8pYbuq2equsTcZONHzirGC25jFX8zg/U+YLKz3OzZQBwbddnOJz13uYpqDq7ktVZ6out7sxf3B1l26QUUjFUtPHtw2iVBMoT7Z8Lq0P+N0rO+SZNvF772KFQsLmIKb9WkrF1Q7MUVaw0g3Z7ieAHskFsRxHWnj0wwzT1N0vhQZ/pmGgDJMgF9oAD6h6Fr24IM8khtBOztnYAViiPXWQYBn6tm8KzAMyLM/QnjvH3Z63t96ogxYeaPv4Amx7TICVBUGhO3LG+F/VNREIKo4gSHOdBi/BdWC+arI+lMS5oh/n1HyHrJAkWU5ajAEv2hdotRuxMciYqViUPIdumDFddvjM1R0GsfaPWm6WyGWM79h8+XaX5VaZQZKRyQLLEHg2JLlCKpitOpxslCjbgm6YszNI2BnHNNzqBsSFopD687Xej9kaxmz2I9pBRifMSHPJxjDmZLPEXF2766gDgz/L0F2aG7tDLm2UeHqpfujCLmVI1RX4loH0LL798VnWexF5oSh5d3J6Jhuwk60S0difa67qYQnFv/ncbV7ZGvDkXJWPnGuxG2QP3SU5mvt1Ya7yhkDk7ejKHCuwHr16KLBTr9cP3f6BH/iBt/Vgjtav//qv853f+Z38uT/35/i93/s9lpaW+Gt/7a/xV/7KXwHgxo0bbG5u8slPfvLQMX74wx/mD//wD+8JdpIkIUn2dzuDweCr+ne8XbU3jmr59+SMPOwc+5DJ4T0kqBrshKR5gWkIBkm25/Q6VfYwDO3Tc2qs6gKYrWrfHoAb7RjXFsxWPTpRuPe756oeL/a0I3MvzjjT8jnV8pmr+fRivTAKdNBjw7V4pROSKclnr3dBad5FIHNsw0GhiZmDMGV7oC8cq92Q//jSGksNn4WaQ823UbIgzAoqjsEwyak4JoaUtHyL7bzAdW08y8CzCqIUSo5gtuIyzAraQYbKx7EHCrIcohySPCM9cKF1TR2Sea8E8+N68Jo4K4MGOhPAUnB/mb9jwnTVYaFis9JLGcY5BdqtWQJhknOq5dOLCyzDomTnRBJGCQzSAts0MU2FbZo6fNQycR0L05Asl20aJYsgk2wMUrIiI1ce5x2LE3UfhWJrmKCkoh/mJFmBaZqcbBjM1AwKWWAJQc2zGSU5Vc+h4Vs8vVSjPYj40u0uaa4dmU/UXRCCmapDkkmWGh62aeCYBkvNEluDWIMhlVJ1TQwBoyjjSjpkFOdkmeT1Gx0eG78uVzZHfHkrYa7icHEu0rESQjBbdZFScm1zxH/48hqg+I7H5/nYuSl822JzENEqHUhVH9cEFGSF4vG5CosNn6pvsztKeGVrwO4o4aW8oF62cUy9NjxMl+RuwOaNgMjb0ZU5VmA9evVQYOdgh+VrUdevX+fnf/7n+dEf/VF+/Md/nOeff56/8Tf+Bo7j8IM/+INsbm4CMDc3d+hxc3Nzez+7W/0P/8P/wN//+3//q3rsX81aOdByvtuu5WFGVklWcLsbEqb5XSWozti1uBukOJbJYt2mUXL3TAcPJrXDfvTFcqvEajciSgs6YULFtVFVSLr7nJ2qZzFfszV/oFWi4pks1Eucny2T5BIBOJbBQr3EN51qEGQFYZLxR9fbnJsqU3VNfMthfRAjC4VCcH6mQi4la52QtV6MbQhMJCfrLklWIFH4tkVeFHSDlH6YkhcK24Kqb3J+yiNVgo1+ov1YsjGPwjUZBBm5oc0DbUNfcIXSyecHwWec61HXseHOW6+jlxiFft3HU8R7OlEnUgeEdqKcMM3JxiPfSfenHYM/iIkKg7zQHRhySabgdjdjoeEx7Vk0Kg6eYTC0DKyyw1LTY7HmcnknRGYpAjCFYhSldCIH05K0yg5PLlhsDVO6o0Sbb5IzX3XoBwndUcTOIKHqWZRc7Qf1+taI8zNlrkU5q92I3VFKaTzm/aEPLxMXkkRC2bP47nPTeLaJbQrSXHJtJySXEtvS4+ZBkuGZJqZQvLzW4/ZusPe6xFLS9C0GScbtbsjvX97Fd/W5/vh8lS/e7LDajVAKruyM+EQ6Mz4fi7GthHHnmiP0pmmlG7E9SmiVXR6fK/PUXI2XVZ+ZskOU5CSGHGd2aY7RgwAI1zKYrbqkuWS26t7Vr+fo87xdXZljBdajVe9ogrKUkmeffZaf/dmfBeD9738/L7/8Mv/iX/wLfvAHf/BNP++P/diP8aM/+qN7tweDASdPnnzLx/u1qCjNub47Ihu7L7+VWbKUUnd1+hGF1JyFoxLUCSG46tvMVBwaJZcgzWiVXE42vfGCcCeB+fxsmZPNEuu9gE6QUhQFvuNworWfuJxJ9hYt24BhnO8tTAc5BOdmSsxUbHIJv39lB982iPOc2bpP3TVBaG5AWiimKw6LjTLdIEcNMxKlyJXg+dt9VroxVVtQ93UmV5zrBn9S6HBJUxSgIk61fLJxxIFUMF226IUZhtBdnRw98hCAPbYHlnIf2yQKrEyfXO+UUM53a93r9XsjNVssYWeYs9XPNSjlTlC0GyoEBakEC4lhglFow8EsL0iVSRjnlGs+hhDYlmAYZbwaZHSjFNe2+MByjZs7I8IMOmFKw7cYxTmPzZV5cr7G52/s8ofXA5Jc0fRNhNDjp1GUUfEtrXiMc67vBjwxV6NR0l2jXpiiJLy+OeLx+TrvPdlgN8j0Bd82cSyDq9sBQZrzoTMtTKFY78Y8vzliN9DHMYhSgqRgob7PXTw/U6XwC/ICmiWba+2AxboejS+3fJ5cqLPSjREonpyvUnFNhnFOJ0johvrdOLjmTNaHUVKw3gsxMJBiiEDwAx89ycZghmvbI0apxLMMMql4/lb34cZL6sj/3B+IHHdljutu9Y4GOwsLCzz55JOHvvfEE0/w7//9vwdgfn4egK2tLRYWFvbus7W1xTPPPHPP53VdF9d17/nzd2pJKfnMlV0+f6NDzbP4k7Pzb3rXspdgfm0XWwhOT5eYq3n04/yQBHV/l6TDGSexEOu9PtujZI9EfLfW8bmZEpfWB3z26i5KwXzNoW4o4h/9hyw1fJ7y3b3fcXlryJdudQF4arFGL8rG+T4R6dgG/2TD4yNnWrx4u8vV3ZAb7ZCqa1H3bRbqHvN1H98x+dCpJiVHEKQZUS5BKTZ6EVmuSA0NjlzDoO4qcqXII0WBJoa2R/r4xTi4spCwPcroBCmp2ueMSMAfd3cKeedF+RjkfP0reoPOWjIGrBL9PrpSA1wh9JgrTQt2TYNCQV4o2mEGqkBJDfDPzpb477/1NP/z797ktc0+Uiqu7QYIAWle8K0XpjFNE6XAMA26UcHpGZNRkrM+iCnFJmemytRci4bv8OXVLt0oZ7HmcbLm8sXbfXIFz9/qcm62yrPLDVY6Ec/f7FBxbIZJyijRm5OKo80DB1GOgWIU56z1EkxTMLdQYeV/+p8RhsAvu3xwyqPimvTinIrr4Dn6HPcciz/x9DwfPttCAFXf5tpOyEY/5MZ2SJBlWEYZ29jn+TmmHvMmWcHJps/tboRtmHTChFxVONEsE6WKrWF8aP140PFSkku2Rwlxpv8/lT9YLuBxV+a4jtY7Gux87GMf4/XXXz/0vcuXL3Pq1ClAk5Xn5+f57d/+7T1wMxgMeO655/iRH/mRr/XhftVrGOdc2w1AaMXQQt17KKLfQXVCkku6oU6JzqRioV7iycXq3kjqoPfOQf+eq9sBWREyjDOisQvrcksTIWfKjm43V3S7eRBlrPdDkqzgynbA7baBEoLGez/Jkws1zmFQQi9oq91oLxRwpuoyX/OBhLJjsTuKuLIdjgmQBhgGRSGJCkU3zKhGOfWyy7ecnxpfXLQyrOnZfGFlQJql+LaBYxu0PJMokxiGQaskCLOCfrTfJ1CAQY41JotIdJyHZRiIAztLBfo+Sl80TfR9v1Gdkt8tdZDnozj8fuUKPBtKtgVCESYFlpSMkgzXMimkJExzVAG5NHCFIs4VQkmUBCUlg0Qb9IWp/kztjhJM08A1BGdbPmdbPte3R9imAC1W5IOnWjTKNl9c6WMKRT+RnGn5LDZL9MKMZslhlOQMk5z1XsDLa0OEUJyerlDzbZq+w+YgQkrJTM3FNWGUSHaChKmSy0yzzM6f/Qs8d7NDd2VInPc4N13hA6dae+f8ZBUxDING2SXJJUlWcH13RJTmBFmGkoKVTsiljSGOaWgA49l7BOjzsxVOT1XoRjqpfTKuOtnyWW75e90oIR58vHRMFD6ut6ve0WDnb//tv803f/M387M/+7N87/d+L5///Of5hV/4BX7hF34B0Bfiv/W3/hY/8zM/w4ULF/ak54uLi/zpP/2nv74H/zbUUYBS9SzOT1cAOD9doerfK1bxzuc5qk6YKLtApzI/sVC5A+hM6uAu6WicxWxVJzJvDWOiVBOYJzN8xxScnSqz2g6wTCikpBvnpIWi4obEaUZWKCquyYmmv5fqPFt1eXy+zKubik6gW+RhmuFZuqVeFHpLbhugpEAprcy6thvimAarvYjpss1szUPRR2JwolXizFSZtW7IyxsBoySFQhGPWzATio0AikLhmnpEZQhwbZuluotrR6x2MhKgbGp33WGUYSIpxg69UaYvmsd0nXdmTbg+6i7f84T+OslzzcUxBJ4tKNkmnm1RcQqiOCPIIMolX1jts/z6NmlR4DsmUhYIpcgVqEJxuxuhpGK56WMbgt0w4999eY2yY7Lc8HEci3OzNZZaJd63VKNQgi/c6jCMMi5vB5hCMF1xWGr4JLnk5dUef3C1zeXtIXNVj4uzFd67WGNzkLDei1gdJDy5WGOx7tEZJby8MaTiWczVSozSDIHgxm5AmkvyQnGiVeJ8UWFl7I9zoulzYbbC1e2AjX7Ita0RN9oBzZLDhbkqK90QyzC43dEcoNVOxGovBhQnmiUcy+BDp1soYKUT8vmbHeJU4jnGnuPyw46XHmYkdWwCeFz3q3c02PngBz/If/gP/4Ef+7Ef46d/+qc5c+YMP/dzP8f3f//3793n7/7dv0sQBPzVv/pX6fV6fPzjH+e3fuu33vUeO/eST37LhSk+dKZJzbcxjAfb5dxLnXC0Y/MgUs0J8FlulTjVKqGA52922Bok7AxjFuo+W4OYNJf0Ik30PD1dZmekU8RnShZPv/x5Fjsenz3b5MYg47G5Kp98cpblps/V7YBulPLK+oC1XkSU5kRpgWcZmELL1Q1gYxBTdS1Othx826LqO2z2I253Iq7ujBBAN4wJ04yKZ/PHn5il6jsg4OrOiCCGfPwnmgY41n5yeTea5CcZNEs2712s048zCiX2LpJRAe1Rgm2ZOCb4vqFlxBZ0wuIY7LwDS3B36boptL+Oa+k7pYX+HLTKhlZKlWwMYTBftQmimGGqKBR0Rim//doOYVqgFNR9PbLKpQbn17ZSbNtktuLgOza9MNUZWp7Nh043+cCpFnN1j/m6x2ovwbO0+kqgWO1E+I7JuZkyjZKFkgVfXBnw+kYfCQyilNu9hPluxEY3ZHcYIwuJb+lzEyU42cw5O13h8WmX7n/8bWqDmODUM7QjiT+WiKMUX7rVZa0XsT2Ima+6bA1jtgYJz93skkuJMATffLbJ6X6Zl9Z6vL41osgl/SQjL/QZEecFrXG8xTDO2RrEtEcpG4OYhZo3JhhrReRDv28PMJJ6WLn5MTD6xqt3NNgB+NSnPsWnPvWpe/5cCMFP//RP89M//dNfw6P66tfd8qwOqrDqJeeuj7vbSXyvVvBkEZmEAt7Pt+JePjznZkpEY3JizbVplW1aJZd2ENMJMuLEZBDneI7F2ZrHxbLiz/5lTQ7/h4/9ESuJyc4o4X0na9RLLr1YRzc8d3WHa7sRhiiouDYCge2YvP9EA4TAMQ3SXNLybd633CKTklFScKujgxFtAzZ6CUoozk55PDZXZX2QUOQFSa4vVrYQYCmUhLpnIYRia1SQFSALmLYFCw2PYSrZGkSESb5HjJXAbgwTqqyTaE8ewzj213mn1r1GjIYC09Tj3CjXnkoAg7DAMXO2BwnDOCNXijhV+JZW3JUswTDOiQuFY8FcvUKYFPhOzs3dEWkOqSxQZeiGKVFSgFRa/WUYzNU93rtU48XVIS+t93AMg06YMggzTrZ8wjjnpbU+W92Ysm/x2uaAnVGCUIqqX2WqZPLCSoffv9rmxm7IyabPuZkKX7jZ5VYn1AadQcJzr2/x7f+X7+UEULq6xu1Y0Cg5eoSVS7JCj5uyQmGZWoYeRCnTFZtelOKPXYrPz5a52QlY7UUIoRPLbcPAMgVPLdY5O+3vrQ1xKmmVbUq2ye4o1oaHg5ipsovvmoeytd6Oehi5+bE78jdmvePBzjdqHQUoCt7wZL7XSfxGrWDHFDR9B6XUHXPxeyWm98cGZjMVm3ag4xumyw7vO9Gg6ln89qsRG4OY5YbPfM0lkwpLwPpwP6IxzguSVLHWlXz6lS0+fHaKmYrD6xt9fv/qLr0gG8tiXZQyWGp6eJbg2eUGYZLxynqfP7rZIcol33pxFoATTZ+8KGiPMmxTabl8zeKz1zvsDmOeu96lH2XkuSIXkBd6VFVIxUzFZZiEDIuxescQzJQ9dgYR3SCjH0smr/hRNVAK+7ro43rXlAF4lv5C5do/aVKmgF6cs9bPkGp/rKmAmqt9fDJpkMkMobQisFFymDJhravVhiVbMEolaSHJixzDUGS5wWtbAz5wqs6/fU5/hn3LYLbmAjoI2BQCwxQEcc7rYUbFMfEMQZzlVDwLx1AYhkE/yugECWmaszmIuNUOCZKM7WHCzihhpuqi/P3z/dx0maArGSYpV7cDClmQZBlKSixDj5+yXGGaBt98rsWLqwOSXPL7V3ZZbPgMowzHNJipODy9VOfCrF5jVjoRf3ijy/YgpurZe4pOxxT83pVd1gcJtzohm4OY01Nlrf4ar2FKaaPDg47tD1sPw+15sz48x92gd3cdg513aB0FKMAbnsxv5iRWSnF5c8hKZ8RMxeXcTOnQibyffZOO/XB8ZivaMHAYZ9zu6g6TZepwUMfUuViOBYt1n4pnMl9vMF2N6QQpo/Z+qMJjs1W8GLphxlovYrUT8p4TdabKNrKQRLlEIKnOlEkKBULx6uaAJIebOwGrXS2ZV5hcmKtyZqbKRaDuWnx5pcMotojznM9e79FPMkwDtoYJQaoQAsqWdkEuFAhDe5TYaIdk3xGUbIub7YBekNEPC4QCzwHfMRmExSHH3uN6d5ZCe+84hSLLORQE6toQZmqv01MozesxDJBCj7jKrsX20KAXpWwMEnphhikUjgHSEizWXAapJE4zigIcS5ALxfYw5b9c2qUfJ3TCgppn8NhMiRvdhOu7Q1pllwtTOnIizgpNgLcEQaoYxOMQ0rLNlS2JEDrj7UzLxxSSTpSRSYmRFZiGQZrvawP/y6vb3IjANk2SrOCF1T7XtwMsS3ButsLzN3v0whTfsVhquLiORVpk/OGNDo2SQ9WxeHqhRrPs0o9zVnsJyy2f7VFClGmk7zu6c1P1LJJccnqqRGeUsDWIAaHd1scihsn686UVrcT8plNNLs5V75q6fr+x/d1iJe5Vb9Zl/vLmkNVexImGz8X56jHgeZfVMdh5B9fRWfUbEfUmJ7FSEQ3f3gsKvV/bNkpzfvPlDV5Y7dMs2Ti2yXtPNPYyaJSUhwI6CymZqzrMVl2iMS/nPScazNUTFusun7na5trOiJJlcXrGJ80hEBlLjRKnp8tsO/sI4dkzLWoDxX+7vMX2MOH69ghDaPfjmaqHFAYn6vp4TcMizXM2ehGvbwdEqeYLOLbAEIr2MGGu4lLkkpfW+nSjHNvWviY32iE6+0qCkgilQY5jmhRASSgqlqHBWDI2rTPAMAT9qKAbpyRSxxUYaA+WoxlM9ysHDaDCY6nWO6bG9kiAHmOF2WFSuQJsE6Zdi7CnCcsWWq2VK4FtCsqOwc4ooRMkZIXEFAZBIemHObkE11JEUoeHIgTC0NEKhRRIJCvtAYXSxOUwETx3s2BzmJONx1y2pfPdrmyF1P2xZYJtkOYCw7BYH6YoYfD+kzWiRJHLgq1RxnzVpRcKmhVPd1tn9uN8VtoBolQhSnN80+DGzoj1fsRU2dGjYVPi2yaZlJxoljAQXNoYgFJ4ptAhoBWXIM3HEvKI2arDTEUrMc/PlDk1VcYdK6+2hjGzFZfvenqeL6z0iLKCkmNxakpvquKsYLUXsTYmOs/WPE5NaXm5lJL/emmbq7sjzk9X+OSTs3cFPJPO0EHRxP3S19+MD0+cFXxppctaL2Z7oDeSvmMdd3veRXUMdt5F9UZEvcnuZuJLc3U7eMMsmSyXdKOMYaJTnzd6EY/N1/bS0293Q9pBwnzNpT1M+HdfXEUpxVTFpebbe8GB52bKjOKM33l9h3aQQRnOzpS5vBUwjAuESHnPYo18sO9vtNwscTMMmSo5tMOUV7cGbI5Sqq7BxfkqpwvJuekyTyw29OIa5fSiDN8x8S2T1qzNU/MVKp7N7V7Eb1zaIoxTCqmYLrucmS2zM4hxbJNhkCKQZLkGOp5j4No6TLEoJGGuyAqYDNmGiWJWSuK0IEo0SpFCx0PIibHgA75vKZAeA513VCk0eFVAfI/3Ji4E9ZLNVCknyqAooOaYFIZBxTYYJhmjJCPKNNBxHEWeFMRyrPjKgUIiDBNT5JQ8m4ZvAYrtQUpPgakEQkoGgaQX6ggWBSRpRpAWDOKcQipKjkWrbLGyG2GJnHNTWo0VZ5KVtqRQCVujhDST9EMdwmsYgg+dalLz9i/CnmVg+SZpCs/farPaCYgyeGLe4emlGlGm2OjHzFQ9nlyooACpJKudiFRKzkxVeN+JGtd3IyAiSgqev9mhKBSWZWg5PUe7zNqIcLlV3ttwTWJlXMvgREMrMUGPoSedlmGcc3V3RHuUAiM+HLf2jE4n4GKykbvdDdkexFQ864HS1x/Wh2f/GdTe7WPuz7urjsHOu7yO7izSQtGN0jHAeOMsmYpn8exyE5SiWXY5Pa13ZfGBGAkhBCXHYhjlXNoYsDWIOdEs8aeemkcqyfO3usSpxLHAt0xqruREw2cUF+wME5RS2IbgSys92tsDnh7/7qs7I17dDLi2G5JmBbNVl0GUMlUqkxoST5jUfJsn5iuYY/v7pm9xqjnAMgRPLdV5Yl633v/b6zusd0OSXOKZgmbZocgLhknBTMnBkBLHErRHKcIw8C3BdNVDKkmcS3xTEOfZnhugDlsUnGp5ZEVIkErSYh+0CEC/osf1bq1JTta9qpCKUVpQchwcS1FIhTAgzzK2IljvgzfWrZc9Qc132U7DQyA4lZI0Vzi2BUKgEIziAtcxMIXEd2yCVJHk2rLAsRUl28C2TS5vBeSFxDAFQih8y2CqZNGNFVujhK1+zNkpn0vrfW7uBmwOY8qOSVForppCMUozbndyJtasRVEw49v80fqQz9/qko+TzXMp+I8vbFJ3bU5M+TiW4NWNEX9wZUc7JMcZp6fKlF2DIJWcny0zU7H59RfWWe/HpIXi6YUqO8OEtX7MiaYed4Mevx9Ufx718bo4X2V5qnQHZ2ffakN3diqueQe4mICqaOxrZJt6bZsECb9daeWubfJNp5rM1jwNyGzzkUxGf5Q7Vcdg5x1WD/Nhu5d/zlFgc7d5tlKKJCu41Q6pl2y++30neGKhgu9qaehKJ9zzvXl6sc752TJXtwOu7Y5gLLt97kab6YqH55jsjhLm6x5CKFxL0AkSXrhd7FniKxRhJlFq/1LwlZUeYSJwTEHDdzg9XWa5VaYbRLy0FpAVOsG5HxV88EyTW+2Q5250KDkGF2cqdMOUX/nCGl9YabPej3Tr3zFYbvnkUvL5W33iTLLYcGlVbKJMj+CEYTBICqQstH+KZ9HwLHphQj/MydBjp26YUfMtHp8v0x7EvNre11nZaEO4YxfBd2/ljB2TufvbKIBhlCKEwDIFQSIZFDoOZC+cVIJrwExFk4sRAhPtyG0I6EcZdc+mUXLwHROUoh/ExLmk7OgAUKkUSZ5ADkUOiZCsdmOqnkXZMZkp+YySnJ3AJMolWSG5tDHkX37mOqeny0ipSHINcCzTxLMFWS7ZGWV87mqbZ5r7cu9b3ZiL5yw2hjEKiW0aPDZfx7UE3SBlra+VX4YRI6WkHWRc2wmYrmjbhlFS8MJqj7mqR5IX3OpEhGnOXNWl5FqkoTYb3R4mPLvcYL6u3Zona9rdwIAQmpR9tAzD4JNPzvLhuEXVs0gLdZg/eGgjBxdmK3uq1e1R8raaEAohuDhX3RvRHXaXfzQMDx/1TtUx2HkH1d0+bMBdwY9SikGU3ZFQfrcdlFLqkI/OBLisdALWezHNkvbsMUxzb46+PUyoeBa9IGN7lOA5Fk8t1VCc4As3O6z2IlzLpD1KMMb8hfVOwOYgIZOSuuew1Czh23qn2Q4SUHBmrsGlH/sZdkcpN4YpMSYl12S5WeKPXZxhumTx//gvW7y2PkQCy9M+a/2I5o7D5a0htzshloCdUYpjGqy0Rzq0M1Ocbnl8x2NzNMoWv/jZm6z2IgzAtwXDKCXKFXXPolk2yPOCzUFOKg1myjZJpkdZZQfyDGplk1EuWRvE+JaFLA73ABT7JoLH9e4tCZQMGDch2In2gU+Qahm6QGGiXYZzDneD0kKPtwZRxnuWakSFouKk9OOCrJBaoKcKar7Ncsvj+k5IphRxAUWSYQwVKE16jgBTQZZrIrxrmVycrTBddbi5m7DRDbEMgSwUo1RSCjN2h5ogPF1zqac20zWXqbL2nLq6HdAJM7o1m1/7S/83gjTH8nVn4uJcFdcy8S2DD56usTvKSQtJxXGYq7nM133STDJVtnlsrkqzZPHsqWniPKc9SknHQb0LdY8oLfhjj81wfqbMle1Aq8AqDiudiM1BxCguDoX8HgoOzgpt6HgPfo1hGHs2Gw6SOJVsDGLKjoVj3l1penHe4lReftu7E0dHX49aBtej2Kk6WMdg5x1Qk26OUurQh+1kyz8UiDlB2hNQtDmIiFO5l1B+1D9nUkc/xHM1l81BxI3dgI1+DJTGi59+/GTHkuaSUZwTZ3IvFuLifJXzMyV+5/Vd3eURsFh3+cLNLrc7ISXXouJYxLY2EXvPYpVXN4e4lsH1XQ3kom/+03TClKptMefbLNQ8ap7NrXbI/+cPt/nK7S45UHYtHpurcqpZYqnqcG1TW/dnufY+KTkGvTBmGBX4jmChUWJpqoRjGlBITKGNAsMkYyeQSCBIMrJCYtsmeV7gWYJBUpAWOf2wIB5v2YuwQAjoJjkd8jvStwWQHgdgPRKVSJDRnTysTO27Ld9v3GUI7aPz/EpfB8eWbHzbYGuUkqYKaetg2/VehGkZpIXudMaFwjIMpJS6eyS0SrDqGZxslXjvqSbTnsULa9pg07O16ss0DFxTEeYSY3yun2qWsE3BYws1TrdKfGmlS1xIlDJ47OQ09t/877m1PeKEa3OzE/H0Yo3TUz7XdwKeu9HDMATf9dQ8Ty3W8BytfLJN7YbeHiVUXIvH58p89nqXjUGMb5dYaPhYpqBVdnlyocrV7YBX1vsUEmarLt0w5dL6kOvtgLNTmiSd5pJulDJbcVEovrzSA+CblptvqHBKC03wXqz7eI6hbxt3ByF3u0h/NUY0j1IG16PWqTpax2Dn61wHuzmzVffQnFtwd2+dCXgZxgVVz9xLKL/XCXz0Q1z1LJolZ5yJ5bHY8PfUEZMF4fxsmeWWz612yPZoPxZie5TQ9B2qvsVyq0SUFkgl6IQZmdKqhW9abjJX9+hFOS+uDVjrxLTDmPVeTMW3iZKcpUYJxxKcapV15s8g5ks3O1zZGhFlkumKy596ap7vfXaJm52IX/nCGq9u9rFNEAgSqUjCjI1+rLOpTBtTwM12xFzFZqnpM0hyXFNgmgJjVOhARyCICkpA1XeYKjlIJemOYooDMQ9vFCK556tzXO/KcsenymQkdTd7pMnba4z/HbVRMoAZH3JhgNLhm1muMFRBkmmOV6YgyCVhkSOlwrcNLKFwHQuBDqkNsxwlFaapvX08x+SjZ1p8+Nw0v/rldbb6MUlRYCKpONbeMVimgVKSYVJQcvQFVwCObfLR8zOcaJVplV2eOVlHCMEHRgmvbgwYJgX9IGUU56x0Qm52QoTSgO3ibJWq53BhrsJio8QT8xVeWJW8tt7nxdU+jgnzVd05kkpxslnmqcUaSS650Q5Y68UIAdtDj1bJRaJ5dgpFxbXphAnDuBibGcq9PLzZWqR5O+Px0NEudpIVSKWYq+k8wIe9GL/VEc2jzGWZ1KPWqTpax2Dn61xHVQsfPNXk1JgkDHf31jkoMW+VDvtK3O2kvNuH+MmFKlmuGCYpC3WdaxOn+T64qbicmtJqhlPTZZRSfOFWl36YkWQFs1W96IRJTl4UnJ0qsz2KeWKuxscuTHG7l5BkCcM4J8xzlhq+7o5IyTffepWF2MP81m/l3FyVz13vsN3PWe8GbAcJjiF434ka71mq8xuvbPPa5ohXNvqM0pwsL/Bsm5IlSGVOPDaCS1IN/F7bHHDTFgjT5PGFOjXXJEgLorjHzqggR6tvRFJgGymJa6KkpB1mJHcBOC76AnfcxHm0KlcP5v8o2M/OOnr/igHnWj5bkSRKUoJcIQVsDeVekjoACdTL4Domjm2QZIpMjlWBlqAXqT0rg0zBMCn48lqfc7NVfNskzAuqrsl81cMyTbKiIEhzpBIEmcQxIMkLtqKMbpQzjHP+T+9Z2OOXpGmO+4efZQboXXw/uUzoKBgkOY5pYgstc98epCg5Iswlca47m8M44zNXtnl1fYhtwqmpMs2yw8vrQzxrxHtPZJybLXO7G9EdpdhjB+YTTZ/lpu78dMOEqbLHEwsVru2ECBHr0OBC0h2lIGCp7u1Jxw+CESklr6wPeGm1hxCC959s8MFTzYc2H3wrI5pHnctysB6lTtXROgY7X+e6o3U4PokPdVjuomA4P1smyYuxxNzY4/dc3hruhfodNOc6+CFWSnFtJ2SUZrTK2kjw6nawL990x/LN3r58EzhgJqglou9ZrPEbL21wuxNhm/Ch0y0en68i0MffKjla4m2ZlBxB2bFwk4g/8Td1tllvq41jGRSF4uX1IYO0wBLadTbMJL/6wjpJJqm6BkqBLCQlx2S5ZbPW04ozQ2gpuBCKnVGCb1m0Bwk7QcrJRpmSY+BYBkutMsoMyXNFLyqIMpCqoOLmSKWf3xP6YjO5SJlA3dc7+M3oa/WJOK6vRT2o0bXiPjYDBtwepvSjgmTM3VLGnfcXChYrNtWyixCC3VFGP8rIChhmCgODsi0plCDKFFIqhnHBi+tDbraHpFmBaxqEuaRsGCjDoFX28G0LYUhutkN2RilV1yIrSW51Q/7oepvTMxUEgp2tNh//9m8HYHarzdLpBtfKDi+v91mouVyYqxBnOVEq2R3FVH0HQ0DFtRnGKYO4YDdIaZVtpso2vmvimoIwK8ilIsslW4MYzzF4erHGB5Yb3O7FfGFFE5k/fGYKx9Jjp/Oz5b0g4d4o4+kTdc7NlDGE4Plb3UNgxLUMXlkf8DuvbbPRj2iVXWZrnv67HhJsvJURzaPOZflGqWOw83Wuu3VdHmQnkRaKXpTtS8xbPklW8MWbHdb72vjqVKuEd0TlcJTYLETGKNEyzYkDqm0ZkECUHZZvnpoqsdqLiLKCnVHKfNVlZ5RyuxsRpRmbw4RLG0M9UkOwUHc5U9Kz+m6Qc6szoi73Gb3/9rkVlk/O4FiCkmPT8B1s02Sp7uLbFlEmkQhOT1dwbIudUcwwyim7DiWrILYUrgVCFAhhstaL8W2LG+2Isg1ZnrIxgNVuiIWi6VqMDEWUS0aJosjRWUOGiQFMlQ1SCUpJ4kyHgwrTpuWb9KOYY7zzjVn3UmrlEjYHBdn4toMG3pMR2WRsqkxY6cUYg4y6b1P3LWLbIMsLGq6JbxvIsTFe3ReYpsXZ6RKb3YDXNgOCqKBtZSw1JXnZJc0VJ1s+jZLBei/l2k5IVhSgFL5tcnm9z84g4X0nEpaaPqtbAR8fH/f/9vwKbq3O6RmfpxbrCAGdIKHs2hiq4AsrfcqOyXtPNjk3U+b6bkirZLPc9FEo1gcxWa4l+N90ssEHTzexTEGUFmwMEkpTFivdiN+7vKvVZ8CpKb2Z2hxENEsO52fKbA+TvfVHoMffMxVNRJ6rejimYDCOwvAdUweX1lxONHwcU4soHlSxOul0v9kRzaPOZflGqWOw8w6oNyIU320ncfAEnPBpbnd1G3ii7j66SB8lNldcg6bvUHHNQ/LNk02f291or6W8Nz6zTU40fFZ7EbMVnXDcKjssNFy2+hAmEiU1N6HsWtzuROyMUlzLZKsfMUxyQrEPdi5tDtgoTD719Dwnmj5hVkVJWKx7pIVklEg+sVTnjz8xze9d6fB7V3dZbpnMV1xcy4LtITXXYDtIySUMw5zYyACJY9rEuWCtN2KznyAENMs2dd/RPIkkRQFRrmj6kGHRKNva48TUzsmdKEOhKDBYbDls9NLjiIhv8PIFzFZ1JMLGoCAcp59MOkAeWqXn2AaeDbuBRCoYJmAK7fvUHkKzJLBNg61RzlTFZbFqUQiDIM6oujZpUXCjPWIQ6dFrkUOSSRxTcGaqzCjJWevFvLbRoz3SBPqSDVlustqLud1PsA04P10iTPbPubVeTBRCyTPJJeRSkeYF17YDbnYCslzy+HyVNJN8caXHbMXlo2enEEKvSyvtEYUS2IaWYgsEf3SjS3uUsFD3MAx44XafjX6E71h85MwUCtgcRLy+OSQtJGkm94DNTMXh9y/vcq0dcG6qzCcuTuPa5h44SjLFxbkKHz7T4mTTp+KafHmlxyjJWKjf2zQQ7j5+ejMdmXczl+UbgWv0oHUMdt6B9SA7iYMn4IRPE+eSuapHs+JwZmy5frAmOVedIKPhm3uEwWs7BudmSns+PIZhcHHOOuQpMbFkl0oTC6WSrHQiLEOw3Czz/qU6X1nrj51QDRQwSlJyaYOniIuCKC0oVff/liQrGEYpN3cDpis2UGMQZdzqRZxu6mR33zFYH2Z851OztCo2/SCjE2ZUPZPpikvVM6kPYjBMyhZca0dEqWS+ZlD34IVRynC81hdFhmeaOJahM7Ek2AbYtolpjFOvCwgyxWzNJS0k/TijF2XMl00uzvq8th0RHwOeb4iyuJOrFSnIckEhDApVHPLo0eEJkGTgFDqWpOwKslyQZJJs7LxdFKBQlD0D0xSEWUYnEsxXHVxf0ItSruwkVG0DV6BHuEBUKFzbouoabPQlm70Bm4OcqNCj1lGiuUJxITGKguu7Aa+s98YkYF1pljPdcGj4Jnku2R0l3OyExGlOO0gxDcHGIKEf71JybJiHZ081cSyTF2532BnatEcpC1MlgjQnzAqGsX6Vap7NTMVhZ5jokVPVYanhglI0fFufT2HGczfbfPz8DM8uNxjEGf9td0Q7yFBK8f5TDRDikADjvUt1tgYJL671ubwx4vLOkKbv8LELsNzyQWivromkfXJRfzvHTw/CZXmnAYtvJK7Rg9Qx2HkH1oPuJCYn4CStHCbGWnfPhXFMQZzKsQFfaTzCkkC0Jwk9eFK4lkGS693kxJfnpVWdb7PW9Vhulaj5Dp5t0PBtvK0Rrm3hOyajJCPNFKYl+eCpGf7w+i69UIchTqrkmORS8Z9e3uDCbIVTzRKZUjQ8mziXRLkmXHaClN1BmctbA4QQVFyTm+2QlU5EISWPL9S5MFOiUXboxZuYpoFlmpQ8F88yMciRaE+U3SCh7jtUfYNMwtmWx0fPTBGkBS9uDNkd6s6QJSCXkiyVRJkEWejbx0DnG6LKQDEGGgf5PQINOrI8I87uNCScuDLnEsIITBRTnmJ5ymalnzFKAQOqrkXVt9kdpUgFVafANhS3ejEr3RClQBUG1RKoRAPzqmvS8ExeWh+wPcoYBDnZgYOzLYvzs1WiHLqBjpt4eW1Ipz3au89Cw6PVcBklevNxdXvIle0RU2WHuarLQqNExTMJE0kmJa2Si+9YnJ8ts9aP+IhrsjlIOdHyOdH0WWlHbAxizk2V+cjpBlmh3dJnqhFRmvNbr2whhOCZk3U+cnqK5262sU2DbphwdVt3fDzToOYalEyTVzcGzNfvdF/eHiVsDRJe3xnqiJs4o+xY3GqHbA1j4lTi2oKpsseTi1UMw3jo8dNbAStfTWBxr+N6o+M95hodrmOw8w6th2HFPyg4mvhULNQ8Kp5Jq+xiGBkNf18SOvH3EbCnzGp4NtvDmH6YcKsT0hmlbA21YqtedmiWHDpBQsWz6SfaTfVWJyaVGRXH4nTL47PXtDT3lX64dzwSxe4ooZCKfphSW6ojDEFbJVRdi16cURSSL9/u8elXNgiSgrOzZS7MVFHsz+1NtIS+4llcnKtxdWfITMXjsbkKV7ZHDJIB8bi7k+WKOC8o2ya2ZVN2HQoEs3Wf6UHKZi8ZS4GlNnob78Z7Yy36sdr8G6MiAHWnv07ZgpIFO8GDKfQKYCcGITJyqQNmSzZ89FyTXqQJ9kkuKRB0wpwwSckypbljskAYgrJj4tqCpxcbLDZcXlrrEyYSyzKYKks6geaXNUral8ox4Y9u9CmkZL2fUCT7R9oLc7JhihwkZDn0opRcKjphyofPTvHfvW+B9UHK1iCmWXJ4crEKwEonOpBcDrujlJOt/fXEtQ1+5/Iu13cCLsyVmSl7vLoxYKMf0yw7NMsO33KuNXZhzmiWXF5aG3edlOLMTJk0V3scxPct1Q9dnOeqHlJKnp6vM0pTTjTLPDFf4UurA3aHCbe7IbZhgDFEoXh6qf5Q46e3Cla+WsDiXsf1IMd7zDU6XMdg5x1cE3+J+zmM3vexR1C/axnM1/w9n4qJm+mkcyOETii+1Q642Q5pD2NMQ3Cr0EqRrCiYr3nEWYFvmXiO9vipehaXN4fc3NVz9ycXa5RcHc5nGYLfeGWbW52IPCvoD4K9YzzV8DAym51hxKlWGccUlGzBZl4QCA2OCgU3t0esDWKEEExXXXZGMVNlm27ZxTR05tWLa30GsaRkC87PVBECglTHQxiyyq3OiGGsiDIoOQXCNDGUYpTkbI1ScqW01BXICkkvkjTLDrZpkBeSYhxweNzY+caou73PNjoJPcrGPksPWAroRvo5DaHNKG93Qhzboll2KNkWnm1QKEUnsEBk5OMk9qoHZcdkedpnueEQJDmOZeCYimbJxTEcSnZGo2RxerpCkEkKZWjnYV/7PvuV/fDdpYYHvsW1nRDHgpMTSwgMumHKzU7MU4u1QyPsOCvYHEQopZBS+wftjFK2h8neemIoyWcu79AJU3ZHCR8528KzLVxHk/87o4Q/uNrBcwymyh5np31eWe+zO0zphprXN1118WyDOJW8sNZnvubvXcQnoOVj5xQvrA4YJRmrPe3UfLutuUbtOGGp4dMNU52RN14zHwR03A2sTDrbD9Lp+WoBi3uBqAcBV+9mrtFXo47Bzju0lFJc3hrypVtd4P4Oo0dR/iQO4ijqn3z4J50bvRDok3KP/yMlv/aVdW53AlbaEaapJagnGh5BmtPwLGYqLjXf4kTDp+bbKKUIknEiuW3SizKeXqozVXXZ6sds9iOyXCIMRTc1+Lk//pdplmw+dH6W/o0hBj5bg5idIGF3mBDEOZlSzJVdkiJncxDRDzNqrk3DM+kHKe0gRRgKy7DIcp3FFWYFaa44PeVTKzncao9ICr3rtSyLggwJ9ANFJ8hRIqdVMql4FjMVh6fnS1zZHrA71K+rG2e4FiSpvrgJNDfiGPB8Y5XJPtAdJfr2/UqgF9aDSSJ74EhBVsCLqwGGBadaJRpTDmv9BKkkMxWX7WHE2HaLmmfy2EKFUVzwB9d7WiKeFdRckynfZjvIyBEk0sA0TG52QpI0Z67q7I2gS47Hb/zA32Sq7DA7VaGTjY02hcCxTJ49NcVukOLaFt1QR0GAVkh5toltwI3tkJc3+0z7Ngt1HYR5slXi/EyZl1cL/r+fv8kXbvXxLYPz0yWW6j4Vz+YDy006YcIoKbjWDpivOkSZ5Oy0VoNt9SPyQnKzE3B6psIHlhu8uNanE+hXb3IRn4CWOCsIs4JRItkeJbxnoUqr7DCMcwZxjmeb9wQbd9sATr7nmOIQWHFM8VCdnjcCFm92RHYvEPWg4OpR9s152DoGO+/QSnLJajc65DB6avpO0vHdMrLmau59Uf8kgmK26h7i97iWQT/MUUqRFopUFkz5LoWU3NgdoTAYujkzFQeFICsUcZpzdTvguZsdRtEE8FicbJX4xPkpLm+P2B7GSKWYqbrsjHz+wyf/Ios1h7/iOAhDYAgdkyEQXNsJxh4fBl0TwKAoJK4BudSOzJaAKJNYpiARkhxF07MoOSbTFYNcKnYGEbujFFNIkBCkxZ5xW8IYsCjohAVppugFCV/ZDWkPEyR6F69Q+KZBdzy8mkQHvFFN3HaPR16PRk2IysX464McHhM9lpIFBAdQ8FFgrD9P+/9iCUUKK52YNCvIx20fwy0QSj+va8GZqQoLDZ9r2yGDOEWiQBgYtklUSJJcYpoGzZLDdNVhlBT0ooyqo12L40xSLnvc/KEfZsU02dyKxnEWgp1himUK6r7m2ZUdk9mqy43dEc/f7GIKwWPzVc5OlwjynJJjEktJq+JwZqbC+Zkywzjn6u6IW+1I/81CUPNtXMfkgws1HFNwaWPIZj/iTMvntc0h3TBDScknLkyTnp/hf39pnTAtGEQJUZoxiHLWeyElu4pjikNA4aChasO32ejHbA/1WP1E02epWdpzg5/UwdDjSUDoxDvs6CbxYEfnYcdS94upOOiSfy9O5b2e815p8cddm4erY7DzDi3XMjjR9MfqJjjR8O9A70opLm8Oud0NKQq1l5FV9ax7ov7JSdyPMm63A27sBpyZKnNhrrIn95wqu0xXHBYqLje7AVmuc3gqjkHFsZEIXNPk5fU+64OYzijBMgyqvkXFs2mULNa6EVmu2B2lTJdtVl2LnWHM6bFK7PH5KrmEpbrP7UIySPJxyrqkkALTMCjZFmXXIp2psjOKxyoWLRcvew5l18AWBq2KQ6tksjnKQUl8x6Rk22wPU3pRQZLmFMU+58Zk/yIkFdzuhQzSjK1BQjoGRAI9rhil8lAnx2Z/xz5Z1o4a1B13fh6tSu7xtYnm7xRjBDMhKys00Gl6kBQ6AqJQkOf6s1KyYThu9aSZZDtIEQosQxEnMIjBMqHum1R9iygpONvyKLsGUmngjtLuy3M1faF7eqlBw7fZGQ3HY46CNC8YRClhVhDnBXXPphtldEcJvm1hmyZRWrA1SkHpC/VcxeE/vbjJH17fpR9mPH/T5RMXZzjbKlFIbXiIgM1+RJLmvLrR5+rOkDhNUShqJZP5msfOKGW5VeLSxoiXbvfIFTw+W6YTZmz1In7z5U0UcKJR4nSrxO1exKX1PpfWh4RZzsW5Gq4tSHO5FyjaKrk8uVjdM1TdGsTsDBMavsXpqRKLDZ/lVhn3AOCYAI3b3ZCtsRxeSj3enkjiB3FOmkuWW/6eL9nbOZY6uOaudkJWu/tmrQ8KeO6VFn/ctXnwOgY77+BabpVYbmpp5d12AnFW8MVbHVY6IcutEt91doF6ydlzWJ5IyQ8+bnISJ1nBlUGs1Qz9mNmqo3cycY5vGTrfqubRuZSxO4xxTUHTt/nEhRbbw5zru0PWexkVVx/X2Zkyc1UPy4Sv3O7TDXO2BjFVz9Z+HoWk7tnM+AZPbl4nbysuZU/TrPmcmirj2RrcdP2Ymu9RcbQPySAueN+JOoZR42Y70iDIMZkq2czVfZoll0GU8OLagI1eQs0zafiC+arHRjkmKwqSRGEIfWGqOIJCCYJIggm+DY4hQCpKliArtOOta0OY6IvUBCAJ9MVqMqawTT2SeFA33uN6tKpAq64SeSfAzYGSa3G26pAryaubMbkCx9QA2xaQKg1qgljtjckm/xwJvmtzczekkAWNkse5mQrzNYdOmHNtZ8S17RBJwalWiWcWy2RS82teXO0zSjLaQUKcFtgi4dz1S3iuRXH2ccplHeNgWyYzFQfPttgNUmxTW0a0g4SNfsT2MCZJcz5rCP7s+0/wAx86ydYw4csrPS5vDukEOmtrlEocAc+cqFP1bHzPYbbicm074L+9vsXtTkjNt2iVTGqeya28IEgLbuyM8G2TNFdsDxP6YYZjCXIJcS5plV2kUqx0Aq7vBORqgELtmRJOTFDLns1j8zVOTZX2gM5EvDABGmGasz1MMERCL7BZ60Qg0COwKGMYZdxqh3tUgbezc3IwWHkYZRpgDo/VUV/rOgY7X+e61xz5gebFSi8S6/0YzzYPeeLcjbMD+1ETdc/ghZUur28P2R7GPHOyvkf260jFdMVhe5hQ9yzWupKs0JbxlzaG2KaBbeoE5t1RygdONbkwV8V3LOKsYL2fEKY5ozjHsw0ESudTpQVTBnzf3/jzAPyPv/JHTC3USHLFWj+i6losNhrMVl2SrOCVjQE1z6LpmeyEGXXP5r0n6jw+X6Ud5szVPM7PlOlHWj2yOUjoRzkX5kzyPMcUYAuJbRmYhvbUaZRdXBPymgYytmkwiAoKKSlZYBoWyXjkZRoKU8F4EoaJ3rELof+Pi7t3cY7KkY/r0a17BcYWQC8qaJZNBALXFOSFIiv2PzcGOh8r4k7ALAR0hjGDRP/MNRPWezHzDR/TgN1hSpBmpEXBS+tD/uFvXeFk06VVcolzHUcRpgVpnmNmOf/4//nDAPzkv/oD+mlBq2QSJjmBbbA8VWK+7nF6uozvWLRKDhXXIs4cMhSFgi+v9qiWHM5Nl7i6EyC7IUEqGcQSIRS2bTFf83jfcotvuTCNIQTP3WgzjFMubw3xLIO5sssnH5vFEIIolQjDwLMtUllQ82wMocNCpysu3/H4rE5S3wlY70WsdkNONH06QcJVxV7H+/3LDZZbJQwh9oDO0dHUBGjMVl0sQ7DejxglOa5l7SnF8kKxPUo4le9TBd6uzskecDoQrHysjvra1zHY+TrWvUDNg86LFdAs2cxWXGarrl4h0bua292QKNVL6MHHT4DQrfaIVErKtolrm+yMUj6w3CCXsDWI+W+vJnr2bpk8OV/lxbU+K50QEDw2XyHOJaYBN9oBpiHwbIuL81Vcy2C26tINtcNqw7d4ZX2AZQiaZYdz0/vKkDiVrHa0s3LdMUlMg7PTZSxT8OJqwGo3IkhyrpUcRlmOKQx2gpQgKaiVbLYHMWtdTci8sRvQD1Omqw5TLvzKlzfYHY09c5TefZsG7AxSXEfgmjZnZkoIIYnyhDiVSAm5UoS5QojxmErtk5MVeieu1L0vchZ6TBFmx+Gh38ilgChTbPYDFmoudd/ENrQvTn88BzUAKQQm6hDYKZvQ8KAT7n+G4rFHVFwo5qoOTd8EYbM9EERZTpgk9MKEimvSLDk4poFtmEhD4hj70PtWJyRyJJuDBMc0kUpR64TMNUqgwLEMnlqs8vrWkLmaS913qPkOjmXSHiVEScZaJ8AyBM+erDNbcWkHCU8t1fgzzyzRGOd/xWlO2TaJE0mhFFGa89r2iG863eJPPLlAJ9TGg4/PlXl9Y0gqY953osl3v2+BkmvjOZZeBwcxUVpQSHBNg1bJpRuluoPkmCy3Sqx247019GTLv2PtPAg0NgcRQVLQD3Ow4FTTx7TMO9zi3+4SQuA5eo08lZePeTZfhzoGO1/HuheoeZB5sVKKlU5IJ8hIpaRVcnAtY+/7W/2IOJOcmyndwdnZHETc3A1JMsmJZolTrRInx6Q52xTkUrE2CGklDiebPkkmWe3FCAUlRzBb0QtaL8xIC8l6P+Z2N+TUVImVTkQ3TGn6DicbHkopqo6JwmCq7PChM429Y9kaRNyMdskKTYierTp0w5SikLy8MRyrqwoSqShyhW0Lpss2r272iXLtUKsX4ZR+kBDmiq1ezH8OYtb6CUmhQYonNG8ik+gZVQYlB262AwwDklwHiZqGZBRrj5Mi3yeYTi44ZQvqvs328KDO5nCZxn50wHF945UvIB6PPqWCYSQxiDEw8CyTQhWY7I9DDQQlRyFT/VmzBZyddrEMg81g3/nYAHwLqq5eG55ZrmMZ8OlLW7x4e0QO5Kk+P4UwOD+rQ0Av74yo5vvLfJwVdPMU2zKZKhv0ooyvrA0obQf0RjGb/YBr2yOyosAxDD5wqsmJZomdYUI/TPjlL63RCTKeWKjyFz54gumqRzfKaZVscqUv3hMVaZLlVH2LpbrPbqhzwXpRxseW6lwQFZ0kr+DMbImybzFbdSj77t7GzLW0WenmMEFKvbtYauiYGiE0ODHGbst7a2jLv2PtPAg05uue3qzkBVFWjLlElUNS+69mHfNsvn51DHa+jnUvUHM/zs2kklyy1ovJlaJkm1jjVGFQbPYjbuyGdMKE2arLxbkqmdQOykpKyv9/9v48xrLszu/EPufc/e1LbJmRkUtlVtZeRRZ3trrVC3vY6tbIbbWlsSF5gB75Dw8koaGWPUIPMAY0ECDZgC3BUAsQBI3sEaBpowcayRa19s5uspvFKhZZrDUr94w94u13v/cc/3Hee7FkZFZWkUUWyfgBiap48dZ49577Pb/fd3FsMqU42wxY6wR87rEuzarpuLx4oU07sPnyuzm7EzPHD5Mc35ZUPZufvLqMtARCSmwp2BklRIWmKBVxVrAxCNkepdzdD3n17gClNYM442I34EzTp1Nx559he5xhVR1KoB/lOJZkL8qxhWn7S2Fh+xZagRNAu+pRcS3e2ZqQFOaz2pYgTguyUqPRCMe052fzJY3hRrRcjNxWSoQ0wYVhXuAKgW0JksKkqmtLYEmBZWtcy3iiUBrwYglIi/JEIDNT7OTKuN2e1o9maQHykOtyrmA31ECJY4HvSgQKB6hYoJVCA0s1y4y2JNQ9h/045/BZvxRI2jUbLSHKczoVx5ChNfjTTqJvg2MLKo4kVQoLQdN36YiDzo4GFqad4IW6xzgpWO9FbCrNvV7IJFOEaYEUguWmx+4oZa1TJVeaV+8NuL0fkSnN7V7ES7cHPLbY4MmVKm9thaZbU/HYGhkVqdaw0qzw3GqLOC+Js4K9ScadXoQQgu1RQst3aAUORalZaR4VYQgheHypxmPdKpujmH6c88rdIWudCp+80MazpRlP1Tz0NJLCfUjgp5gqxTpVj71JRpyVfHt9OB/Bf1TroxZD8YNaH91v+EegHkSCO4lzAxw54I+rtdbaBx2cmucwSHKUhlv7E752q0dWKJJc0ZsY1cRaK2AYpdzrR1zb8XjxgoNlWVxdrrNc99gPTZp5t+JgScFSQ3G+U+FTF1vc7BmDv4vd6lTyntMLM75yfY9v3xuxPUnxLEm76kzVIwVprlE64da+xZnp5z/fqfDYWpdX10eoUhFmBRcXKqAFq00PoTV3+iGDuMTBplSC3iRjkhXEWUngWnQDD1vIeUdFoElygeeUFJnZEVsSci2oODYLDZ/FwOJbmxPiXIMDRa6IChMn4SqN1uaC5XqSTs1hkmT0I01aGsPBk6piwag8IJie1o9mZcoch77hvFNgjgcB6BLCWM3B8qQEuwTLgpYrObdY5e3tkGv7ESiNKw1frOZAq2IRZVDokkkS8z997S5V32GcKCqOpOJZLFSND832JGV9mPD4cp2qB35x0ElYaPg8dr7N5y53eXd7wjfuDtiepKBNKKgQEKaKqmehlRk/3ZuOxHthRuBZuNqMfzaHCfcGCdd2AjaGMa4ludgtWW74rLYCAD6+1uLiQhWtNV+50SPKCm7tR9gCbu5HbAxiAseYna62zWMOe98gBJ+61ObWvkdvkpLkZrS1XDfmp+uDhLNNj3ZgusLv7oQPDfw0AKrKxW6F7alRqeCjCyhO862+e3UKdr7PdbyteZJvzvlOwJ2pN87hA/7qcp0Lncp9DssfW2uyNUy4tT/BlRav3O4jhSAvFVGusKXk6RWbu/2EV+4O+L239/gvPrXGn33hrHFerbi8eL5NUmh8S3CuW+H5cy0udCusD43jaafi8dSZGp5jcacXcqc35PrehDu9mG5gg4Bu1TiiBq7N7f0J47TkG3d6fG76WS90A37s8QVsS/L65oA4g8faHguNgLe2Jnzj7pBRYmSxcZEzzhSdikOnYpF5Fp2KzWNLLQLHiMXf2BiRpAVFXlL1LQpV4tgSWxgvn2GcUyqNxKcoTefGleBbgjDThoCsQEhzkUoKWLAFuDb7UU6pwH3AOhOeSrJOiwPi8WpDEBYwiDSpBk9A1YHhIevlcvqPEm70c24OBvOwUAtzf8eCXMPGIKdgqgSUEKU5QtpUXIFrSwLP2D6MkohRUqC05t3tMQs1l0wdCOaLQrHWrvLEUpU/erfH7V7EJCnpVG2ajoXSkrZfslD1cV2bO4OIflyw0vBYbQUs1jxavsMLa02+uT4iL+H2/gTXstgcJni2ZK1d4Rc/dvaIilRrzblWwMu3e+SlUVqFacEoztgdaxSC9YGxpphJzeOsxJKC1ZbPZy+22Rql7ExS4qzkq9f3eGt7YkbuNY/FhkepQIiTOY6HwYzv2nzmsS73+jHn2gGuLT+ygCKd8S/z+/mX76c+qmDue1mnYOcjVDMUvzWKSTI1983RcCK3ZzaLPn4gW5bFn3luhd1xwjfvDri1H5MrxaWFGpOkAAFL9YBJus8gyvAsi3d3xoziHN+1cS1BoSFMC5Rrcc61+dhaG9eWfP12f5pfk5Er47y8XHfZGSVkpWKx6jJKc2ypWKi5XOhW6EcZm33B1iBilB+s9llR8s27Q97eHvOte0OKEm7sh3QqzrztnReaZmBT9SyaFY/Vpku32ub6XoSQmqor+dylLp2qxc29CRuDjGFkzAS7VYd2xaMdSG7ux4S5IilKxlGG1JqsAEuYuAzLMnQeLczFxBbm3zjOGSYlajoJSB8gszrFOqc1KwtYbvjsxSWjNMNX0KkYMrBTKIoTWn/JMRfKkqmtgToEijBAqgZoSyCmnlS2NOd9mBlzwrzQ08eXjJOc/uSA+5MUJV+9uc/1nTG39k0AcCNwqHsWlxdrDOMcpt0OAby1OUEIzcVulWfPNtgPczKlp4KAglu9ENeWfPJ8m/MtM4rfHidcXKzNQ4pnMvDznYA/uaHZGCQopVnr+AyijEBpzrdNsOhsrdubpHzjTh/PllRdh+fONVnrVHj+bIN/9+1tbvXMuHy5YQjSGo1rWzy+VDvRj+w4mLm6XJ/zdA774Bz32/mgpZRinBTUfRsp738/D3JyPn7b7f1o3r0/6bM9Sp12h0ydgp2PUM1OunFSUvcPcqdmc2ngPsLySVERs7yrQVywHxovnOdWuzx9tk5eajSGv7MbJozTAktKnjrTYHMYszFMWax59KOUwLVN8nHVO/I+tNa0AxfXMn4UjYrLJy52WO7H1F3J776zwzgtubkXIi1Jmhd8/faAzXFCRRX8u//1/4EoU9ybKHb1hLe3JwyTkjTXiMTY4dd9i8V6gGsJzncCSq1pVz2uLjdoV2x6cclr9wbsDDNeXx9iW5KdoVFvZaXxLWm7moWaR7dqcXM/QhUwzjWWyAlcm0yDIwV7UUk2HUE5QK0CFzo1+nHG7ihjkh8YxZ2CmtM6XCfZDOTAN+/GBL4huudAPy3ndwyk4eYkU2rZSfhZAt2aRT82hLFyeic1fX4LxULVx3ckjmMhhAVoFmomU86zBPnUMkE7Dv/kp/4Sni2JtWR/knB3PyLKDC/oTN1ntRuQ5Yoo1/TCBN+WjNOSQZihhSZKS+OjZUvSUnNnkLI7Me7PSmlqLmyMC769OeLZMy1+7LIZgd3pRXOl02rLAJO3t8dsjGJePNfiqZU6dd/m85cX5rzBhYrD6/f63OuFlErQrjqcbfk4lqAVGGdoWwrONn2eXKkxSUvqgUPFtefuyYfBw4OEILMOiWsJWoHD3f2QsRBH/HY+SCml+K03dnh3b8KVhRpfeHppDnhOAh7AA1W5O5P0iPLsg7yn0/RzU6dg5/tUDwrqPExYrvv2nLuzVPf45IX2feaChw9krWPSwtjFtwOXXpRSDxwCx+Lx5RqWZWEdOsa/+MwZPnupa1iOwL9/fYf1YcxqK+DZs02eWBF0Kh5PrlR5fWPE9ihhteXTClz2w4TXNzTPnG0gpeTqcp3znQo3dkb0QgMSqo7FKMrM7jLOSHON5dh85X//14jzkq7rEGYFnmNR9S2ENinPtoArCzVWmx6bo5xeVJCXinrgkRQlvhPQCmw8W7I3SRglJUWh8B2T9RNmijjX2HHBJElZHxgL+myqkplEmnrHZOFkeUmYlHOejQaUgqpnMYqF+Xn6u5mK5rROC6bBoNJEPxxv1oRAmBz8nE0B88yvqeJI6rYZK4E5BWfdHICmB+faPhUvJ8s1O+OcaKou9FxJs+KQFgVhLuhUBb/4/DJJqXl9Y4TQEik1Vc/BswXv7MI//+IvU3EFVcfBtw23J3AtpBCsdqpYAuq+zSApqPk2UZpTak3FlZQY/sz6MOaJ5TordReNpuNZ/E4/ZpLG/JvXFIVSuI7NJM345t0Ru2FKb2LWILNRcuhUXSquxBOCjUFEUmi6NY/Hluo0q2YjdX0v5K3tMVmhqfkWgSPZHMbEqUJrYwSYlwpbCi52a7iOZHeSmY3gtJv0ztaYe4OYc61g2n02EROzTdqs5vzIUUKhoF2x7/Pbeb81i9DYn2TAhM8kHZpTYcZJtiBwcuf+4HpgNrkf9P2cpp+bOgU734d6UFvxOGH5KCKHC90qcOAOOiMqzw7kGcAZJ6XpvkzdlGeLwPGSUtKqelzbnnCnZ6IiilKhtebKUhUhTTbN6+sj/uU37hGmOU+faXCuVeHd3QjFCIHgmdXGnOj39VsDNgYJxdQt1BJwpuHzxFKdG/sRng1JkjPKSsOvwURGfGKtwSgpuLk3IfBcXFtiOzaWZUCSnhr57YxSOoHDhU6VvUmCbwnezUOy0hAsO77NXphSKBgmRuU1iEsm2cEFySheLJ480+Tm7pg0L1CJySTS0632tzcnoBXpVGUugMCFqob+g5Xnp/UjUjZGnWdbhkT8XuVJ4/U0AzyeJWhXHOI8I8o0RaZRyoxOAQoNN3sJq62Aet2i5mfsjTM8W7DSqtLwJW9tTdib5EySkq/eGnCpG2AJQbvmUvdNp2d7mOLZkppjIQWsNgxQKT2L/UmB51nkRU6MxbNnqpxpeqRFyR/d6CEtTdP3OdcOGEbG+fduL+K5Zxv4js3Ld/YplHnf13cndKoudaAVuLyx0ed2P6biWOSlYmIXbI19PnWxzc445U4vYmOUEtgWWWmUpRcXaiSZcYcWWlD1HJ5fbVBxbSqezd4kZWecUPdslho+pYJhWvDJlTYXF2rzNTHOCl6502d9kLAzMuBhFjHRi1Le3ZFHuiczkYUjBRXPPgIIPgjXpe7bXFmoAaazU/MMuHQtYTpdJ4ylHqTK/W64OJ/maJk6BTvfo3qUtiocJSwfR+QPSuKdHciuJXh3RyLE0ZHWww7w2XuJs2JK2hV0qx6eYyGlNOqOUcz2KKEXZixUXeq+w/X9Ce3AYT9MSAtz0sZZwfY0obxQmvVBzCgreGyhxk88scjHkpw/vraDeuNNiiRneOEyOUYO2oskpdYsN2tkRcF+lBPmClfIaVSFyxMrNW7uRXzz3gAFrDV9klyxVuRsDjMTVBolOBoyYaz5o1yRleWRUYGW0K35hpeUaxqBg+tYDJKMKIEwhzIt5345LiaUseJYDOPyNPn8tHCm4CV5BOBrAe0K7EzMzxqouoJ2ILnXV0TTbo4LSGGk5LmCOCnZ6UeMAgcXWKy7LDd8Li3VudQJ6IXGKVmh2Bom9OKcKM6JsxzbtsimqeWyLHhivMEoynhpcJZSGL+oQktGcY7Qms9e7mBJENLCt21eXGuS5Mae4uefX+bfvrbFl761xY29kPWh6ZiUynSohpHCFoIkL1is+UzihNe2JuyHOZcWKjyxUkdjui/Pn23w3GqTojSCAceRnGkYvo5rCW6NErSGVtXlp55scb4T8MbmmLwocYTkzc0xDd/mqZUGjeAAmGSlnq+xs071bEAogKzUDOJ8yjc8WHNdS5Bkis1RyuWFKp+71EY8ZOR0fER20roqpeQLTy/xmaRDzbO4vhuxPU5oBQ7bo4Sabx8ZuQEPlct/t1ycfxRHV4frFOx8D+okXs2jtBUf3uk5RlSeHsjHTxr/2FMfP1Fnjse3dkPCvKAdOAgxk7lDnBozsMWa6bYs1gx/57FuBQV0qwcg7M7+BK0157s+k1TTDmyy0rye79i8eL7Fvbu7/J//zn8JwF/4v/5bxpZPXGj6YcZqq4LrCJYbNXbHCVKaxbTqm9bzzijlbj9ib5KR5IrlhsveJMMVFtKSZHmBEJLA17jajMMWKo4JMCwLxsVUFePZPLlcZzfMkEIjLGmyr4SFkCVCGKJyMV0zc6BmQZiUJ+YgPahsjEmc1pC8571P6welBMz5MI9SJbA9Mc7aojCAplCwExbzYwzMcdawwXcFWWxiGsJMMchStIamJ3FdB7ef8rFzLf704wvshzlxXhg+iyuIEk2YanSWkytFWmpWbcX/7b8359xn/ta/ZGK7hGmKbUscCTtjuNtL8N2cqmtzpuXz2EKVb9wdMc5KhlFJkpbsjjPiokRgvHpqgctPPrHExjBhoz9hLypJspyX740ZpzkV18aWkuV6QFwqM5oPHC4t1kzyuTZj8KvLdeOYnJesDxIeW6zi2hafudjm25tj6oHDzihlnOWESTEVFIg5p/HdnZDNoVGV1XyLlUbAx8+3WGqYrtSsq33SmpuVGt+VnG0G+I7kxl489wu6vFi5b731HlG5NVO1Jnl5JHi5UOBYRkl7uNv+vQQjP6rKrFOw8z2ok0DKo7YVH9bpOQkkze5/WAXxXplb5zsV7uyH3N4f8/rWmLyEcy2f67shL98bUHckjiUQWqOFYK1dwXdt2hWXp8/Wp8nEIdd3J2yOc+qey1PLPo3AYT8qjIFZzWV7lCIPfdTHl5sMpGMUUmmJbQueWm7w7Nkqv/XOHlvDmFJKLKGZpIpelIEGpTQV1+LGrjEXdIWm5Vusx5oSzUrdp1tz2BmnjNIci5KGOsCj5AABAABJREFUa3bYBVBzJWmRc7sX0o9zvGmUxSjK8C1wbCi1YJLoudw3LkA9IAvrQaUwsuFTns8PT80CYY8fB+/V7csxHUMw/ju7owLPNh1DkZlIEjBu3nGu59yeUhmllgQCSzMIc2quTaEg8FyqnjPfhASuRIxypNTkynRbGp6DIw/aT7XAJhMWstQErsmo8l2bwLPoRzn9KGeYZkgtWB8mFFrx+uaQW70QKRVaKRxLsFL3+Llnz/L4co2buxP+4e9eZxLHpFLQbXgs1CoIIelUXaqexbOLTbzpxuzxpRrLdW/uoyOEQCll1qZhjBCCJ1YaNCvuPNcqcHIKz+aeiolzE+1S981mamsU8/r6kOt7IZcXzJr2qYudI6MtrTXnO8ZE9TDv0bONx48QhqS8Hya8sx2SqxFKd1msufNcrdmGc2sUsz/JSPOStU5AcIIidlZHQkCFoF0xI7lHIRt/GKDkR1mZdQp2vgd1Ekj5IEj+UWevhyXsnYrH02frSPlwVcJKM6AeuLQShdKKL7+7x7fXR+xHGQ3PxnMldc9hkhacaZpWumsJk56+PWG9H7E+NCqOUilyLXjqTIvHFgLu9hNevTvg+l7IoHfQ4/gzzyzREx7v7oegFGdaPnf3I/75n/S4048oS0NetG0TUXG9l9P0LFy7QpJlvL2VkeQlFcei5lpkpZGi11zJOCnNwp2UOBLKqXmbQBA4kru9jHv9hCRXJFKwWINmYDPRGVEJjtA4lnmcxvAzAtfszHP1aADm1GDwh69sTARdckxC9Sjf8/HjoSyMH8/ScoU3N0KS0rh9z+5TwBGpli3NhkMKTVEWKKUJXAudKTo1m7wA37ZQCAOIipKiVHjOwabolz97nq/smBTw3iTHakiagUWpoOHb+K5FVijSXFEqRX+S8dbmiLv9BNuyaFYFj3WrLDcCHl+pU/EcFho+gWezWPMMZzDOudyucH6pjiWNK/qlKaiZEYI3hxHjuGA0nQPWPIeb+xMC2+LSYpXlujdf79Y6AXd6EXd7Eb5jGdGFZ81H9O3AZXfqiLw7SWkH7nxtnZkTnmTSCkfXVNcSvLExJldjHCnmfmIc+h5cSxCnJd9aH9L0bc61A64u1x8avHxSCOh7rf0fFij5UVZmnYKd70F9NwliDwJJxzlBW6OYt7cm5GqERvPsavMI6FqqeaDNnBvgymIVrZb49saQQml2RzFxlqMKxaXVCgs1j7sD48uxOUp4YsXlne0xf/TuHjd7Eat1j8e6VUqluN1L8G3BtzcGbIxi+mHO9iid+uYcwIT/8OYuT15e5ieuLOBIwct3B7y5FXKvHzIIEwokaWnGTlprilzTbLkINI60+cN390ly4yciSUiVQiWKcVqyOc4YxSbaQUlAG5Di2qC0YJwVuAIirShLwcYww7E0sTIGgxKoedNxVmn4P3VPEE2MdN/CKGai7P4L32n98JZrw+QRg89soB0YiXmSm+7O4SqAvVgzTEOwDGdHcLJBZdMFz7EZJjnRdkn7+j4rDZ92xWaSZLy+PsKSgrwokUIRZoq0gKpQTA4Ri/bjnJ958gz9MObVe2Ncy6SFX1qoENgW7ZpLkpa8sz3i7a2crUHMKEpxLIvz7cAQkoVgkBTzOIuFmsdzZ5r0xjFRWnJzN2Y/zPms0lzs1nhipTHvQs/WptfXR7y9M8GVRlqe5MbhvepZnGn6fP12n7V2hasrJsphpvacSdln6xfAlaUqlxZqVFyL5UbAlSUj5JiBhcPCjYdd4IUQPH22jkabfL+KcWVOcjVXaIGJnTHSfs36IOFM8/7wUf/4iOp9hoB+WKDkUZVZP4yjrlOw8z2qR+nkPOwAe6/fzTo57YrLUys1qq5NXpY4lqQfZVPejHVkp/S1Wz2a0+ytQZyz3PD5xY+vEqUZ//j3b5CVmm7d4yefWOTKQpV/9c0tdsYpr94Zcqbh8yc39vnyO3uEeYEuq/yZp8/g2AIpTHL6+jCj4kiWGi5N3+Fsq4m3eJCN9fbWGLtZo1CadtWlLEuEUESZuZpUHYlrQZik9KKcQZhybXdk5KRLNZbqDnf6ZrTVjxWOBNezSKcLsm0ZR2RLmtGBb0Gn5vDEUpV7w5h+lJOVgNAMiox21aYozQihxFykxFQFNkhhmB6kU8/GGeUp0PmRqqQ42RfnpCqAfjwL/DRgJjt2nzA3vC5Hmq5BxbWJJvmR1wgkrDYceklGkkNRKvbGGVXPAQzvZGeU4jsCrSAtFWEyi08pKA6lnr96Z8C51UXOdmqUWnBzf8Lt/ZC3t8a8sNbkSVnlpdt93twcsTPOkMLYCy7V5dwMseJIFqoOoyTHcywsy+K/+PQ5fBf+P19fZ3ucEGWKKFO0K84R7xu0puU7bE8S4jQH12KxViXM1NRZ2WU/zNjbCdkdp5zvVghce75+XuhUWGsH3O3HvHS7z3Ld5/JihY+tNdkZ+yY2x7GOuNArpahN/1bv5VP2+HKNZ1ebRzpCM8HH7HFr02BUgHPtgLpvPzIH81EBy4clF3+Ujfdx6f534jn0UapTsPMRqQeZA84O8oepAtB62skZk5eKO/sRVU9yoVOdE/YOyxlnIXxvbY64O4iRWnCuY3JpzrcD/vjmgJt7CZNMcaZp8dbWhK1Ryo29CaOkIHCNG6x53xDYgigt+dK3N0kKxeXFKv0oY38cM7IkSikW64qlRpOfvnTQQvZdSZzlvLk5QmvNMCkI04LFukdeOFQ9mzgriQvFzjg1u0qtiTKXcVLwqYtd0rLHJMkplElBr9iSJCvxJTRqDlop4lLjWpJaYPPCuRZ5UfLOdsgsvNwGAg9812bNtXl3N6HEAKTFqkWYlkzSo0nmJdBLH/3Cd1o/HPV+XQcOHzMuxpvn+HO40+uYlAIpNR0fRom5n43J27rZNz47EvAsze4kwbMFSa4oCoXSEGWKUhsp+CyPq+JZtIKDC+zeJOHa9pinzzVxLMF+WLA+SNFK8cc39vija3vshwlRprGkOScW6i7tqsvGMCHJSmqeTT/K+Z9fvseVpTo//vgCnmPRrvhc6AbUfAuUZrMfsTmo4liGk/PGxpj9MKEylcK7jsVqK+DPPrfCICm5N5Wq3+qFHFZSwdH1sRU49KNsarFhus2DOGe57rPW9uf3SzJFzZOkuQZyulWzrj7Ip+wkw8HDwGB2/9ltAuY8pAcFNx/O+XovZezh+jDl4u8FupK8vE+6/1EOSn3U+sH/BD8kddwcMCsU/dgYZZ3vHG2TrrX9aTs4YXeSsVQzQXi5Mq55t3ohq60KnarD86vNOev/8InXrrgkhSLPFQiIs5J2xSUtFHd6EbXAQgiHSws19icpd3oh46Rgqe6yVPfwHRNlcb7rE2Ylwyhja5zi2RbjOGdvkjFOCmzLAhL2o5ydSY4IfX5x+pmfP1tD1Dze3JowiAp8SzNOS/JC4bsWT51psDfJGIQpFdeQI31XmrDCvKAb2JxpeNxR4JUFYVoyjHOEgKWGQzPw2RompGlOYAs6gUdeavYmGfmhq1DVgW7doVv3mYTJ/AKVKpCULFZtFAXDg4gh8/f8cA+J0/oBqVni/XtVzv3HTCCgXbVJ84JxqshKRcUyXcjyULDsbLRVYgJrJ0nB3X5CoTRVzyZwJZuDZC6FtyQ0XHj2bJ0179B7FXBvGJMrw2+r2OBamkQZY87dSU4yVYy1Kjadqk+n4hCmBePYqMfu9GKivKRb9dkPc5TWLNV9PEfyuce6lErxrXsj9ic5dwcxozjn3iDhy9d2GUQZVdcAhNWmj2NJvrk+5nynQpor3t4eU7Etnj1T5+JiDdc29hdofWh91HQqHkJktAKHXpQyigvW+zHrw5h+mFH3HRqBzZMrdd7aGjNOSqTMjPrqUJPkvToohwUfD+LQnBTcPNuMXtueGH7SNDriTLPyyPyb75dc/BBUO/bzD3b9QIGdv/f3/h6/9mu/xq/8yq/wD/7BPwAgSRL+5t/8m/zGb/wGaZryxS9+kX/0j/4Ry8vL3983+z7r8Ek3O4FnM+a1TjD/3WLN5Q/e2ePtnTFKaS4vmU7JJ863EEIY35vcyCmX6z7b43Q+49aYGfO5VsBTKzWyXPH29giA5861ePpMnbRQXF40z3m+XeFzj7X4H//4LjuTjEbg8IkLHS4v1RFSUvFtnjvXZqMfMvBtNoYJFztVAkeyH2ZooBVIVloB9wYZFdfm1a2I2s//ZSwpSIXD1n5MzbEoXE3Dtyl0giMlzYrNU2fqWMCXXt/CkZJWzeOZszUUkpbncHOQsNqq4Dk2u+OYrIwZZ+YiEaSKOE/YHudoBf2yxLUTBnFOp+LQrtr0ogJbwlrbZ7lTpeVZ/P7m6Mj3EmcwjIu5ueBpndasLAwocW3oHwPCJ8VInASONaC0JteCQmm0hhTznCUGdGggLw5GYAWwH5pQW0tKSkuy2nC500/m8RNNG5pVn+dWW6gs4//3s/87wqSglym2bvSxbIlnC2yhKRX4thl7e7aN72gudirUKx6TtGBzlLHc8Gj6FlGh0UoxSQocK2e16XNjZ8L1nTGebbE5TAkcM8p2XUnDtxFAP8oQCHphhi89ulWXbt3j+l7Et+4NeHdnTJzmjDKNFLDY9EHDtZ3JfP1aqh9E5lxZqpIVCqU1d/uSvDTGgHlpZm2Bazra3apLu2IGiA8CM4/SQXkYh2bGQ+qF+dR6w/xudvsbGyNu7Edc7k7Vrx9xUrDnWLx4oX2fdP8HvX5gwM5LL73EP/7H/5jnn3/+yO1/42/8Db70pS/xm7/5mzSbTf7aX/tr/Pk//+f5oz/6o+/TO/1gdb854MGs+DDXZhRn/PbehEFk5vqOZUBN4No8s9ogyatmtjR1NH7pdp9RXJDmJRuDhJ1xyvbQdI7CvODZ1aZx8nSs+e7kQqfCZy91qPk246RguWE6SSsNj2fP1lltV5HSSDbBOKa6NvTDAiEUr94d0q56ZLniUxe6PHm2xjBRVBzJm1tjfuuX/09M0pxRPzWSc0uwUHM503RpVWwGcY5nSV7fGOPawuwEGwG2BY5tM4oKhKf58cfahFnJ6xtDslIxSgriIkeV0B+X1NwSC4gxAKgfZfiuzZmmZ8Z2MkFi5PT7k4wwkYhDehkb6CePtms/rR+9KgE0xOn9v3O4n59zUuXaBOJKzDir1NPUc8tiKXB44kwNWyi2Rzmbgwnbk6kdgobAglJIPEuzMQxxBBTCAC3HsefeMbnt85t/8a+xO87QSYouNaM4x7IkgWuCRLWU+J5FoyJo+i4/98wSo7Tk9Y0RaaGQUvDsWpskV0zSAt+WPLlSo1P1+ePbfeqepOE7WFKgtKbu27SrHk+daVAPHFYaAUop+pOUN7cnLDU8nj7bYH+S8tUbQ0Zxhi0l3hSkaK1ZHxrlZpKbc/KTF9rz8M5ZN6UXGXXT5x/rcKcfszNO52Rmz5a8uxPOVVXHR1izOqmDcqIfWc2b5wMeBk0zY8KNoRnDzeIoPFvSrrgooB04KBTtivuB+TfH39OHRSIWQhwJSv1h4OvADwjYmUwm/KW/9Jf4J//kn/B3/s7fmd8+HA75p//0n/Iv/sW/4Kd/+qcB+Gf/7J/x1FNP8cd//Md89rOf/X695Q9UDzMH1FpztxezOYyo2jbUTH7UTzy+gD8l8M3uMyMqP32mftAt8p0pqU6Tl4p7g4ii1NOUX0MgPLw7ubBQ5fpuZJ6r6hClHte2x9zuxXzmUoc/fXWRK0tV1joBTD19vn67zyhWNAOHF1abnG1X+MUXViiRRvkFrHUC3tgcsdELeflORqk0pVK0Apv1QcokzdFY2KIk7CeU2uwyH1vwSHPN25sj7kxn+194apmffWqRe4OY2iRjtV2h6uds7EcU2qhmLAm2MkTjrNBc6Pp0qi6j2HB8hnFBkitcK6Xuu1Rci2ZRMsnBsSA+Nck5rYdUeILmXPJgbo/D1JSQGYHY/I+c5rZpTI5WqUpqrk2SFTQCh1ag2J/YQI7EmBLuxQVSFfQFBqw4EikUtjQk5p1Jwn98c4ezLZ+KY/PEsscg9kjzkt1xii0Ftanc3JGSTs0lTDKksNkYxSzWAi50K/QmKaOsoOJY/Lnnlrm2G7E5SlhuBUitWWt6CCm41K0xjDPu9mOyUlMPbAO2lFnPWhWbb94dgBCMkoLX1kdsDRO2RilKKaqe4HzD5/GlGlXP4VwrAIHp7Ey7OjOg8PrGiC+/u4cjDSewXXW4sljlQrc658dkh4KVIWWcBHNvnkex7tgeG9XqhW4F15YH85xjD5sZE55p+PiunI/KhBA8c7YBGrZGMSuNgGfONj4QeJhxnnpRyspUcXY4N/F858Bg9rtRP4yOyz8QYOev/tW/yi/8wi/whS984QjYefnll8nznC984Qvz25588knOnz/PV7/61R84sDOrkxD7rI06SRWXlio8dabBQs2bp+nO7jMnKiuNwEgpZ90ixxa0Kg6g6U1SNoYJlhDsjBI+vtYizsr57mQ2Ix/GOTvDhJ1JzL1BgmtLSmU6R5cWqqBNiF2UFry1OWaUFnzyfJPFmstemHN9L0JKyc44ZaHq8M76gHdfeh2hBeeWV6h5NgWgEZRo+lFBWmasNn0maU5WKqQI+KVPrPLW5phX7uyzMYhxbIt//eo637jTR0rIS4UqFS3PJq057Ec5eW5k57acXkimERK+BUVZEOcFWQl5obAsGCcJlhRoAZ5lHJRn4Z825v9P2MSf1mkdqQd57lgcGFvaTI9LYQwrHQsqLqSlkanL6X2zEu4NU3qTjFFa4tmGOO/ZRvqsSkBC1ZYsNjzyIqcX5qhSkWQFgzBnFKY8r0a8uNbCf+Yinmtzay8izAqagcNqOzCqRF3yP7+yyc5wwNY45pc+vsrVlTrXtiHsldzphfzeNdgcmSTutzeG3O4nCAE/+9QK/9nTi7x8e8TOOGF3UjKKJ5xrBTjTC3+n4rDWqTGICyquhSUFQmgWas7UM8jGsy1Wmv48tsGzJRc6FW7vR3x9qr5a6wT0whShNWmec2075PWNERe7Nb74zBLv7kRzoLJUN+GiSaZ49W5/PuJfaQQP5M7M1tphlHOvF3FvELNc9+lFKUmu2B4lLE95i940sHNmTHh8VCaEwHUkUkpc54N1SbTWR8AdwHLDm6/P93oR9/oxa+1H5wP9KNZHHuz8xm/8Bq+88govvfTSfb/b2trCdV1ardaR25eXl9na2nrgc6ZpSpoeXLZGo9ED7/u9LqUUr2+M6EfZkRNyxunR2nRtjgOduawzcMiVxpGSXpSSlbU5wU5MtyQ7o5RJWrA+iHEtidLQqcXYUsx3J0IIluoeN3cn3BvEWALaVYeq59CuuuRFyc29EFsKJmnJ7f0Jk8zY1gP83rV9bu1FnGt7fPJCh3FW8u27Pb786m3+xd/8zwH4n37rmwxFnUFs+D11z2EY5yw5HisNjyQrGaUF+3HB9jAFBaM4I80hzkvSpKQXZtR8hwsdH0sKolwZh+TSYi8rkXrqUisFUghGibGzDxPDeXAtYzZorHgESgscSyCFwpeCMtdMCnMBOz7O8qVRysx25Kd1Wg+rkumIaZqUPp3OYGPAdbciCXNBxdGA5EI34NJSlZu7EdtDI6EuS+P31AxscgWTJEdakrWOz8fXmvz2G9tEuTkaJ1mB0BInS/h//Hf/GwD+u3/+h6ycWaSXlGgBaVjQacCbGyN2Jyl3exFZURJmMb/3zh4//+wKF9oBX7/dZxAV3NiPWar75EoT2Ra9SUapNTujmK/fHvLK3T6jxAgWBIJv3htwZbnOk8tV/s1r26z3Jzy/2uQLTy7wh9d7fHN9RCNw+OzFLq2KTVyA70je3JowSXM6VY8ri1V2JinDKCcrFKtNl0lSorRmuR6wOQx5dzfi+u7E2Ecopt0cM/paqXu8dLtPL8zYHKWcbQYP5c4ccT1OcuLcBIjOuD9xWvJvX9tECMGLF9pcXa4/kPeTTpWkSW7+e6H7/tPU00LRjzIcKciVolPx5nL3bBq4HGcmlmL2mT5qPjkfhffzkQY7d+/e5Vd+5Vf4T//pP+H7/nftef/u3/27/O2//be/a8/33SqtNW9sjKcI3hwQh/OvZqQ8k9wb3sf63xrFNAOHP3W5yyDOj0jOZ7uVe72Yu/0Ypc3oSAjJUsPnYqeCkIKdcTpPST/fqXC3F5HmilwpvvjMMmvtgP0w5+U7AyTG66PhSwIp2S1KulMgtt7f5uZ+yH4YEzgWe5OMG7shu+MDB+V3d0O8hnF8LRQ8MZVvrg8TXl8fkGQleamRSvH//uottgYxW+NsDjoKbWIdpGXSiqNck6cZk1QRZ6UBKAosBbbU5CUM45xSgRSShmdhWRLHtqm7FsMkZRiX1D1JxXEYZYqJyilPIO144mg+kuDkGIHTOq3DVWAiIPSx2yYpCKnQWlBzHZaaLs2KyzgpsQRUrJKdqf1BmWgUOY8v1qDlIbBYbHqkpSYqNEqDbRvqnhTGCXlWX7u5z7OWR6fugxYEnoXQilGcsjmM8R1BXgq0ENzYDfmPb+3w/NkmzYozPR8VdU/yE5cXsKXgS1GBJaAfF3zt5j6jtKTuWVzoVtgZp6bbPIh4a3PMv/7mOpY0a9HnL3dwbIuVpofWsFD3WGlWpg7MBa9v7DGMClpVB600Ddfixs4YWwp+952C2/shUkgW6w57Y4t7g2QeIfPCuQ5C5PMuy539aO6Lc7lbJfCsR8oknPmRzdbEK0tVxknBS7d6bAwNJWCp4c8BzMOA00lqr0cFALPOETCP6JFSnujMPOMzfZQiIT4q7+cjDXZefvlldnZ2ePHFF+e3lWXJH/zBH/AP/+E/5D/8h/9AlmUMBoMj3Z3t7W1WVlYe+Ly/9mu/xq/+6q/Ofx6NRqytrX0on+FhdfxgT6dAxpGSXN1PZstKTT/OjiT3erZJLt4cRryzPSFXmh+/vMCnL3bmHhBgTphOxSPXI1bbPuM4x7YEVxbr/PSTiwSeg9aalYZP3T84LM61zUm2VPNwHck7uxHDKKcXptztxYDm/NRkKy0V7a7DubbP2VbAxjBBaHhra0yalwyj9Ihfw83tMU4qef5sE9uyjNW7Z7PVC9kfJ1iWcVPt1D1e3xgZae0h1qdnw5Uln8VawE8/tUiYKq7tjHn17nBO4C4AVYCH4e9M0pKiBM8pqbo2AvP3vbJc5c6eptQ5SVZS9aThPhwCOhIzAhPCqGQsxTyx+vDFa2rYjGtNfVLK9+/Pclrf+zpJQfVh1EmvoTDKP8/SREVBXti8sT6cAnaT0zajj+VAL4G3dydcXW5QCxzGScEoyimVwrHBkYKa76BKTSAOXnEYldwZJkjLEJMFmpdv9Xh7a0xaaJbrPleXbW73EtJCMZ7mZV1eqNEKPIQuubLU4OmpsGG1XWF9EDEIc+4OEsZJznNnu1xdrvLynSG2FLSrPm9ubpNmBVujlKJUPLVS57GlOrtj40q81q3y+FKNcRLw6t0+AkE/yWlVHF5bHxDnJbf2Is40PXbHKa4tycrSuD8vVri4F5CVGqWlISJPR2BJXhpOn2cyqT7/WGf+u/fKJJw5Nx8m6jYC5z5TwYcRjh+k9no/AOBBz/EgZ+ZZ+Oh7uS9/r7otH5WIio802PmZn/kZXnvttSO3/fIv/zJPPvkkf+tv/S3W1tZwHIff/u3f5pd+6ZcAePvtt7lz5w6f+9znHvi8nufhed4Df/9h1mGvm+PeDIcR/CzT6qRQudkuYZY2vjmM6IcFWVHi2jb96VgoycsjxlczK/TtYcz6IKHu2zQqDtIyB97s/SzWXPJCmxl1prAtY0S4M07ZDzPu7odoJdiZZDQ9m29tDEmmJmSL9ZTXNyc8e7ZBxZa8vj0izxVJqVjtVDljH3TotsKcLA8ZxTlrnQBHCv71zR7rQ5PyvFy3cWSAJQVr7YCNgeEHoMEX0K5avLDaJC4FYapIC+O/8Ym1JllesDNMKYCKBYWQSCEYThnHjgWebRMXJVoJtvsxdwcZ/aTAk4JJkhFmJbk6CHnUmN2yZ4E3I4PGMDoB8FRtWGm67I2zE+3/T+ujV9+PUWTLm3Z6tFmMcwUWZjScFSXDtMS1DOn18PvTwDgx/lhPnAm4sx+zPYootaBiC853Ah5baJBrDZPJ/HGeIxlHGddLkwW3oTKUUowThWNLXNvm6eUGrapPWShWWgErzYB21SHLS3ZGKTXf4l4/5kKngmtbDKKC7UlKkhestnyqnsXFhRqrrQpKa1xL8I3bDpYlsC1BmOR85Wafz11e4Bc/vnpkjar7Nt2qz9NntSEb24KNQcy7OyFbo4RhkrPW8imnBqPXdyYs1H0+cb5NqeBj54yvmJSmw3GnF7EzMt3kWdJ49oi25w8CAxe6FdY6AVII3Gk8z/sFDO8HALwXKDlOJn4U9+XjrvvPnG0coUV8N+vDcoN+v/WRBjv1ep1nn332yG3VapVutzu//a/8lb/Cr/7qr9LpdGg0Gvz1v/7X+dznPveRJCcfRvMPymt5mOfDcYQ/IyS/sx2SFQVnWwHtikM7cLm1N+HVu0OA+VxZSmnUATBvMc8OvjkpL865uTNhb5KakEHgTDPAsU1WzL1eRKk0zYrDZy+2iTOz89ubZFOPC02clUzSAtuxsKSkVrO5WvV4YqXOmnuw0JSFJhOKvUmO1jBOcu70UzI1645IhkkJMuexboXdScbWKGEUZsaVFYtvrY9pVz3yUtELM5QyYKQdOAgM2bjtSwot2J/kxKlBHi3fZqXhcW8YIzUMksKAQyFAaMaxIi5MMOPsHWvMWMxzbFzbjMJKff+MyxGw2nIolGLwKPrj0/qRKwkENrxwtsG7+zFJXjKJ1RRUl/iuzXLdRcsCR0h822i85snpmK7lzjimE1i4iwHr/ZC0UGgkC7UKP/fcWbKyJB4egB20JswKpIBmYCFQaF3iOxKEplm10VLw7Nkmi3WPT11okynNm5sj7vRivnarR5QUXFqsMIqMaeB6PwYUo7igH5tzsNSa3iRjY2g2T4En+fTFNi/d7BMXJZMk494g4blz/nwUn2QFt/cjelFKp+rxwqqPY0ve2ppwqxexVPdxbcFK02d3nFEo2BpnrLQq/PxzZ7g3SOhH2XzEP+PL1AOHwLE43wkeGNh58Od58GYUuM/l/lGe76QOjmuZlHWt9XtmVL3fEdCjeAedJGZ5ZvWDKcXeqz5MN+j3Ux9psPMo9ff//t9HSskv/dIvHTEV/CjWUZdkTbviHpktw3tL/g7/fj6aUiNc26Jdcaj7LjvjhPVBzM4oBcGRuXJ2aAGouDYXuhXAnFRLdY80K9geJ+yNElzX5tMX2zQCh07F49JCwNYwpjdOuLYzpuXbdGs+lxer3NgzicStiovrWJRxwSgpqHsOCzXX7KpsidYHA51W4FCrelRci7xURLmRgwPYlsngaQYOUkKj4vL4YpWi1JSFscV3hCArIC1KRrEgTAo2hgkKjSOM1f1i3UMpjSyMn0/Dlyw2A14816QX5sSFQitNqTVSKFwpjW9GoWc5g0fKsyAuFVGpUYUm0/ePP0oNaVayNT5l8Pyo14Jnuoj9FJJDHT45/bc9TsjL0hjkYUZUqoAozakvBlR9m3GqKApJmOZHOGECSAvN1jBhtR1QcyU1x3jmCMucF0leMIkPeHKuJRgVJX7VpepaNCsuu+OMWgGuFJypu3i2xV6Y0a15bI8zdiYpk9SQjpOsYJKW3NoN2Q8zHMsiTMx5JNFIyybKC97cHLPeT1gfRKy1A55dbfHcagvftnhrZ8xSPZhn9nm25Nr2hDu9kI1BTDOwefl2HzUl415cCHhxrYVnC27tRYySgk7Vw7UFUkrWOibOYBDn9434TUfBGBEixH3dFO9QZwaYdztqrsN4yv+b3Zfp9zWMDVF6ue4+sDtzOMrnpNd8FP8f+OAjoPe6jsw8gA6LWWb5iYfruzXq+ihI2X/gwM7v/d7vHfnZ931+/dd/nV//9V///ryh91HH23mH868+yIF0PKW3U/HoRSlxobClYKnh41hiPlfWWhsy26G2rjtdaLbHCYtVQ4qcJDn9JGfJsVhrBwgh2R7FWNJ0bV5dH9OLU2qOzePLYFsB7ZqPb0tevNBmrR3w7s6E8o5iuWbMw8ZJzh9e7zHe6/OJ6fs/26nwxGPLPL5U5V+8dI+6X1LWbZKioO75PHW2yU8+sUhaKHKlCZw6YaZI8oIoN233ZsXmUrdCVkCYZlgCslzjeRaBYyHQ9CcJ/TgnzTXdisXKlH80nC7gSgji2LRgfHt6IbIkjlao8sAczsGAsLQwzORcgTQb4vsIp7dH6lShdVrk2nQDi0NAx5NQ9SRSK27sZRw+9WdEdwvFRi/CcyVxJqh4hjQsMUG1FuA74No2b2xOGGeKbs0Y2BWlJkkL3t4c8HvX9tjc6POXD78AFpO0RCnBaKrkMZ1SQc1zmeQlWa5ZH8TG6bzUDCYpUZYTpQV5qRiEOXtTl/TFqs3+pMCyBCsNi+WaZ7ysopxOxaXmO3z8fIvz7WDqFm02EjXXxrXEvMtwcy9kc5iwXLW5tjUmyQsc26biL5rncR1yFWJLm0uLVT621prLv4H7RiXHOwqz+2gd0wocHHm0U3Ou7XOnF3JzNyRTigudCo3AObIZXap73OtFjKKM2/suizV3/ryzNTbNyzlp+Ljz8+EuugFm90dYHK6TqAtJXn5XwMczZxvG1Xrq3XO8u/RRIRZ/t+oHDuz8INdJ7bzZQT47STQcMYd6L2QtpTyW0isBM09/bKGCJeV8Hp7kJTsT45Hh2YJWYJMeIrNlhcmfsqQJ9nSags1Bws7ERE7c2nPpViwaniQtJFIaCXg/yrm8VKPi2qzUPe72Y15fH7E7Tlise1Q9izu9kG/cHkCW8OWf/Qt0qy4vPrbIC1cWudjx+dqtIY4Q1FyLe4OYflpwYy/kxXMNvnJrYGIu2r65GAhJYMPH11o8tlhlnGt825At0bA9SlltB5xteCRZwas3E8JpiGKUKPajHrf7IVFWMElKwzjWYFvGQ2iUafLCXBdmQMcGmp4ZH2QFTBX26Kkia0ZKlhiJ8SnQOS2ASXa/Si9VUMYGDCum0nMbarZgmGpyDWEGqIyKLxkmUC8kgS0JRYnQBng3AhuJIsoK3tka0K24BK5DlmfkCl65M+BuPyEu4J9//BeouhLbc7CUETuEWcFq02WUaIpcUHUludKMp4G8cV6wVPPwXYuNUcLbG2N2Rgm+Z1MWpRk3S8koNVl2ni15bKGKkIKkULx4oTXl21V5+kyd1zfHvL455vZeiGNL3twcUgscHl+q0al4ZMWApm+6SvthilLw+LJPnBZs5gpbpLi2Ra403apnPG6ma5vWmrVOwPlOcESYcbyjcFjR+ubmZE4l0DqeuszHbI4SzkyFGi+ca1H37fkaPFOo3twN+cMb+/ypywt88kJ7/hrXtifc7ZsNZd03SeuHnZ8P24g8CoflJGf97xb4kFLyzGrjgdeXjwqx+LtVp2Dne1wPsiZ/Z3vMK7f7ALx4vs3VlTpwctr5w57zylKVtCjpRxnrA+vIY2YnmVKKmzsR13YmXF6ocWHBtGgXayZNvOE7XFk2r785ShhEZvFZ70d4jqQEnlxusNRwp3wZqLg2SsHX7wzYGMSsD2I2hwmlhpt7IRu9kChJSUr4X/6r/4b/9s88yQujnH6UsTk04abtis3OyIzgSgWbw4R/9c1NdiYFUV6yN0l55myTtbYPU8l8t17BS3IsC3rTVrZxpdW4rs2bWyPi0gCQEnNxSUvohTmO1BRT1rFtCSyhp+ASjnOKC2A/PXC9lVM0MxvKzb6VUy7yaR2vWZjnjOgO93s2tQIbzxbEKkcVphM0KGFSKCwJSklcG9q+MPlRQFEUpqOZlNiWxdZwQtW3aQcuWZkjdAGqJJMO//0X/2vONm0TzKsUQgik0CSFphO4OLKkKBU39kOuLtWI0pwwLdifpHzxmSW+cafPIC3JtCBAI6TEQlD3LZ5cqeO7xi15uRbQi3IEOe3cYbVdwXUkb25O+P23d9kcxOyME4pSI4VgpVXhQqfCpa7PV6Tkne2xiaipG1fmn31yhQJNkhvPm3PtgNVWgGOJucngjDuzNYpPFHbMSmsD5A64kumcSjDjULarLiBYbfmcaVao+/YRgHFlqcpKI+CdnQmOFAzibDpOPBhZxblZBQLXmsf9HH4/75fDMlvf30tl9UFGTg8bL31UiMXfrToFO9+nOnxgpoXiXt8ABBAs1iNWmj6uLd83ss5Kfd/sevaY2UnWDCxevTtgGBcgQj59qc1Kw2dzGPP25pC4KKcmfRatioMtJaXWbBYpYaZYqrusNI2J3yAq+OSFFhe6Vb61PmR/YtQdi1UXWwpWai639yNevTcmLRUCiLKSm3sJcanohTmbo4SVhke37vOZi22GScGrdwcUecE4zim06Tg50iZNMy4tN7jQ8hCWw3o/xHdtVqoeUV4Q5SYRupsbk8VcCaacYxxhXGpty6Lm2uRljjcFPIYsDY4t0A/oyxwmKh82ErQwwY15ceqz86NaFgeKvcPlCoinN2rA2NIdfVzbNxYLgedQqgmDpGAwVQyp0phXRpnxs5lkpjMkAJ1BxQakJC9KkBKtBYFrsVR3udPTtKoenp2R5YpUCUqlsBDkpWKSlsR5ylrbR0rYHmbERYmlSm4PM5IsJ84UT67UsKSgW3WpehaXO1WQiig34b1/7rkzZKXmjY0hvSjFlhIFrA8SNAJbmpBTzzIdH6UNj8m1BYs1l1v7Idd2JmwMY2qujUCw1gmoeza5LkkLk7XlWhJr2lLdmRyIO5Yb3pRsOyFXIzSaZ1ebR7vj09HS1ihmkpTUfGseu5CVet4VF0JwdanO+U4AQswjJw6vwYepA8sN/8SR1eNLtYdGOMw6PO8HnBwGH0s1z0TwaH3Ea+27OXL6qBCLv1t1Cna+D3X8wLyyVOVcO2BnlKC1pigVX7/VY7UdmIMa7ju4H1QnofHDwAqgH+Vz8u2lbsCdXszuJGWjH/Pm1phJWrBU83nxfJO9MOfKYpXrOy5h2qNQmgudKrYU3NyLuLUf0o9SfuG5M8RZybfWB6RZSbNi0w4ckIIbu2NGSYFAGLfi3T2uvQHPPP84aV6y1vIZxcbf5kzD4y98/CyrdY9vb42I04LHqw79qGBnkhAWRkp+s5dyY3+fwLF5frVBt9qgXfWouhZZIelFCW9ulORFSWCDXULVgZVOhVbgsjsK6YUKLQRCGyO2DNBKz3kTAA3P2Penx1CMI0FM+TplaYDO6ejqR7cqFoyPEZAbjiGrz9o4AuPPdPhAkcAkL9kZZzxesXFsgXXo9xrD8fFtmCQHjt1i+tiKZyMRuBUXR2pqnstjSzWqrsPOOCXOBWkB1dEQLUB1OgSBg0QQZYp6RVILXJqBwyDKGU6VlVleEqUlSo357be2WWoEPHO2yZmGxxefWeLVeyPe3hpTcS3e3J7w9TsDtgYRFdfm85e7rLUDbvdT0rxgkhaEcY5G8/FzDaKsxtYo4dKCUTi9crvPnV7I9ijGd2yeX23yc88u8Ttv7fJbb+5S8Sy++NQZhCiZpIpBnNOpeAhhkszrvk274hJnBYFrzYnPMyfh2WhpexgzSQsKpfnMpS5Xlkyg8YxKcNK46CTOjRBiTh3QWk8zAc2XfHxk9aCadfPv9WPOtQOuLp/cjTpcx80OX5p2tmbKsw9j5PRBQNlHtU7BzvehTjowry7XudCpEGcF//71baOmGqf84sfOstYJuL4b8rVbvYdmusD9aByOjsLWOgE745S1ToAtoOE6/NH1PWwpcC2JLcWc2JvlxqI+yhT1wOKFtTY1z+KTF9pc34145c6AOFfc3ov41r0BUsAoKrg3iMjykopnXEWrvoPnCJ48W2e8N+Af/eqfBeClV69ze2BQhBSmxfybG0P6YUquFHlplvVcQ7PiEGYl28OY17dsClWSFgAFNd9hHKdc6laxr2q+tTE0uTbDFKU0riMIXCiRjOKCQsF6PycqQaDpVgW2kFiWRV4UpLnZffo2BLYgSfV9iisLONNwQCtu9cv5WEIAgYDoFPn8UFfL5YitQFjer8pzLBglh37GWBnMSmII7lrD7iRjFGWUAiiNLL0ojImlVsYpvERjTZ/XkVD3LSxpsVBzWGtV6NZcNkYpcWqOcUtadOsSL5nw7//+XwLgZ/4v/wtnl5usNFyGSU6hJa6EXAnqgYNrC8ZxwV5eUKIJ04JX7o74masOZ1s+n7nUpVULqPoxq50q6/2IjWHCYGIiESqewLMtMgVKlUghubU75t4gRWnF0ytNrq7UaVRc2lWX3allRS/MyUvFY12fS4sVKq7D1ijldi+iFbiM05QzrSpS5veJO8CYKLYqLo4ULDcORi6ztTbOSwqlSQpF4NiEWXGEGHy80344FPlTFzv3AZjZ+GcmHZ9tLh81jDPNS1653Z+qZhMudCr47ntfjoU48Dw7rvD6MEZOP0wk5VOw832okw7MmRvmwVo4bcEC13fDIyFws4M7yUuzUk5PvJNOxFGcszWK5y3ftbZPlBR8486ANC8JXIln27SrLp++2KRTdfnarX3e2RlzpxfjTuXfF7s1ujWXlUZAq+rx8cBha7jE12/tsztJ+Lff2kILjSUEthAMC4UQ0AxcHutWOHdlkbonefvGwW7j33x7izdGmoprQNYkMaTIcWICOh0B0jat5Ipvk5clri3ZGkWsNAMWag6NwOH69ojfeTOhUIpnzlQ52/DYGMT0ohxHgOvYBDYkWUlaSCxbYdmgp6RlW0jqgUWYlmg0rm26NZ6EONPToFIDcCTmQtOqObSqLmmhca2I/JCxYPw+gY4rTAfglO/zg1OOPMrBOT6+VECU3X/bcel44EKSGZftmIORaN0V1BseYZaBkGS56SIgIHCg5kiEFER5ST+GUke8sxdOlYQWj3Ul3ZqLZQkuus35ay7XfXJlAIBWRomz0Q9Zafo8c6ZN3ZO8uj7EcST9SQoIemHKtd2Yj533uTDtGKw0AsOBiTJu7k6MuudMnasrdZqBw3o/QmnYG8cMkoK7vZjVVkCYFzyxUuN8t8ruJGOp7rFYcxlEGaVSbIcZYaZwbcmZZkDLn1CqkjTXPLVSY5IeKJE8W8y7K7thRqfqErgm5uY4TxHgymKVvDAu9IfVR7M8wl6Y0q36PLlSJcnUPBTZs+UDDfdOGvU8CnfmpHX+/cRHvJfy7IGv+z55PT9MJOVTsPN9qIcdmL5j8eL5NkuNmHMt49I5D4ErTcSBI+GdrTEv3+6xM05Zqnt84kKHqysHrdDDDplJpqj71vyk3x4n3NoPGUY5Z5o+T5/1+OylBa4uV9kcJtiWZBAW5E6BTjV136HqWbxwrkUjcBBCYFkWP//8GZ4/V+fXf/cGb2/3sYAnz9Q406xMd1KKi90an77U4fGlmklKrx4sGnmhSfKSOC9IsgLXsiiUoiwVWa5JAauAvXFCO3BoVz2EKElKwSROWai6FHnBxjhjc5hiW5qv3zHS9EGcYwkQlqThGffkSBfUfaMY6biCW/2YtIBcaW7uZuSYpcedEjD606zYmWLGleA6RrWlEPSjkkIbf5TDu/qHYZ1Zivr8WMCQnU+5Ph+tmp2RD/ouh8nJ39ns+xUYmwLrUH7aSZEh3cCh2XXZHUfsjo2s3LEFZ5o+jy/X2QsLbuyEYGuyaeyJwOJ8u0o2TfPuJZkB6ErT8B0udqs8sVLFEpKqa7G1sTd/vXrFY6/QCGmxO47pRQVxVqBKeGK55PJyg16cmXBg30UDkzQnLc1RO+ssXFmqmuDONOcPrxVMspxGUOfJlSav3hswiksqnqTqO/iOxJECz5LsjlN+961dPn2xw3NnjKOxYwniTCPlHraU1DzjB3Z1qcpb20aCPkpy3tgY8/qmCW3++PkWaFgfJqy2/Pm4f9ZdOWwOeL4TsNYJ5hfpwxf7WR7hH1zbZRhltCouadHFcw5CkR8kDT8MHA776zxKJ+T4Oj/zGnrY44zTdUHdtx8cH3HM5+eDRlTM6oeJpHwKdr5P9SAWvBDCZJ0sVOceFMsNH601k6Rkkua8sTlme5Rwp2dayEmhWG7GnO9WjsxYZ14Odd/EKriOZTpBU9Kg51h0aw4/dnmBF9YaXNuZ8Pb2hEGUUfUtnlmpkZaQlYpsShI8fHJIKenWfLo1B9cSRFmJJQRPnWlQ9y2GsaLmSbaGKcOkMITAxdr88c+da5K6KTf3J4RKIyy4tFgjjA0Ruh9nFCVEKYySHFVCu+ZSFgXrcUk/Kqh6FjZQdcGxbYqiJEoy0qykaguqvqRU0E9ybCmxhKTl29zZ00hp4TqgtDJmbpiLlTMllc4uWrNPXChQGVi2wrYUhYZxmOHZ5sHF1G/nQdJzB3hq2WN/lLIem9sqApoB7EXf6RF1Wt+tsjBjpI4P65OTO24nGWMLjCWB1FBxoOLbuFbBbnL0PjMpeglsj3O6NYfAcZAyIyshSTX3Bimtqs+ZusfuOGESQ+locj31WNYlK80qd3sRVcei6ggudJtcWayy1jXZTZtD09G1kwPt1/Pnmrw90VRdAapkZxSDgmEc8/rGiH6UIYVkrR3g2gLPstkcRVjSYrFueDHOlNNypxdyfSdkcxyTFoqNUUw/SpHC+F999lKXwJF8/c6QlUZJWpRc35vw8u0+m8OEs83AmKFWHSwBF7s16r5FWmj+3be3SYuShmfR8GxKLVifxtyAplVx2BmlbI4SttsBv/ixs6w0D3L9Dm/0fFceGf8fXndneYQSQT8uaFc8JmlOt+ojpTxygT/OfTwJODxqJ+TwOn94vX7Q45RS/NYbO7y7N+HKQo0vPL300HiJ7+S9HX+fPywk5VOw8z2o99s6nAGW2QG7VPN44VyLb60P50ZUy3V/zvZfqnusNv15Qu9S3WOtHcyJdUt1j81hYnZBTZ9PX2oDmp1RwmLdpx445KVmfZCQlQrXslisuVxZbjJJc8JcnbjD0Vpzb5CwXA945myDvUlKWmp2xykLtTp3+2Mcy0KjOdsy8vZ8VPDM9PGPL9W4HmmGqTdvFbcDC8/2YZygtI1Ek+aCMNNICvKxoihKJqnZyUZZgWtLVKlAFmhdkqQlepqIXmjjYRJ4JrriXLfCvV7MfpSRFcafpxZYhsvDNOzTYZ6JVagD5VUGWBqyHMq8QE47Pbo0IaO2AFuaC90wMY9NDxkOlsDtfkpgH4xAMg229WAF2Gl978sCWr7g6nKNUTFhkBjOViANQfj+gBBTEnMceA5UXRstBIF9tJfnT8dfmTKdnrCAV+9FR8aYqQaRKm7vTQjTgqrn4NiSlrIZpya7TVg2RalwHYusKHGkpF21cR2b/UnO5ijlra0JSsGSPHjHzcDh+YZPxZEURcntnsc4yRHS5m4vIswUV5crXF1pcWsvYpwWCCF55kydt7fGvLsbcqlTYT9MuTuISbOSs62AUZyz2gxQGjSai90aHzvfIisUt/sxcVYwigWWlERZwbWdCf2ooF1xeGtLEbiSC50qn7nU5lvrI9aHZjew1PBZbgZc7FZAwyAy/bGzTZ83N0fc60WGULwbMkyKOS9xe5zQC3M2hjFnGiaS4rhj8mydnY3kzncq1HyLM80KlxcrTNJyvsE7DiBmr3EcOHzQlPP36qCMk4J39ybsTzJgwmeSDs2Ke+LzPgjUfNAuzUfB/fi7Uadg50OuD0rwOnrAmvC55YZPXsYs1T3OdypcWaqitZ4T9b5+Z8AwzrnXiwzLvxXwqQttlNb861c3jLy94fHnP36WTnWFb90bkhaa3UnGha5RhK33feJc0Qhcbu5OSEpF4Fgsn2ncd3IkecndXkTNt7i8WMd1LHqTjFu9iF6U0YtyKo7NixfM+KtdcenvhfPHf/naHm/v54RpPl0wM272EuKswNKaUaLQaHzHJi1Kwlwhy5IiNyOCrFQIJRinOUqDFCVlaQidctptSXNFzff42HKFx5eqhGnBVwcxhVK4tqBZcag6ksApjHS4hFF2AHJsDjx64EBenE5/UMVUdixAKDMCy0rz+t40sHScHnj8DDPzj0PP10/0d5SKfpg7clrfeWXA5kQzSsZgmY6cFIZMbGnTwZvV7BJgkqtMJETdM07HRQlYAl9AoqcKrcBCK42VK9TUzyk/5sA9A9mTtORsSxK4giizcC3BQt0zHd/SOB5bQtCqeAgBm8Mcy0qNIZ4rjfJLaqr+wYXq+l7IQGTkpaY3TskLY2QaJhmFYyEEBG6DtVbAa+tDNvspvifYHefshxm2FOyMEgZhxiQtONvy+fHHumyFGSt1n0lastLw8B1Jmpe8uxvy1uaYYZKxXHHwLChUSTtwjG9Q1SbOFMV0FlzzHdY6pjOVl5pPXWxxccGEJGeF4ny3ghQCpRS/+9YuYDY9O6MEE3uXcL4TTJ2SNRXHwnflkeDk2QbyQreCdyiP0LXEIRn6UfO++wDE9DWOA4cHdUJOUuAe9+85yRRxVnXf5spCDTCdnVkH66Try4NAzQ9Tl+aD1CnY+ZDrgxK87rMJtyVoc3Df3jMdnOWGD5ojHg9ZoRhPLeC3xwkrTR9HQlaU7IcpcVbye+/sUfdttBY0ggPjq6vLdc63A67tTLjXj3h7a0yhNMuNYBobcXDiJnnJrb0Jr90bcrsXsdb2efFcm7uDCNey2JkkVF2Ls02fn7q6gGVZKKW4ER6AnY1RyigumWQ59cQiypXpyJSKUVKSaY0njMi24mlEZlRStmtyfBabAbuTHCkM/6fUei7N9TTkSlFFcnXB5/OXu6Ql3N7vTf++Nq2qQ8Oz6YcmtBStsBxpumcYY2WL6WjCml7wMMTTjGkXaApuAkeS5IqshEkxDXp0oO1ZOJaiH+kTOwIlMP4OkI7DyVyQH8V6L9B3XC31sCqBUcFBG0cbdZRrgVUegF+BSS7vpfO7YVsSy3IodUGhBa2qZBwrXMuMTKNUkypzbDVd4zczyQw4Ko3eANeChZpLWip6Q0WalzR8m49daPHs2SbDqGBnnAIarRRb45TAkaw1fTNSGiagBYs1j+BQZ2d3nLJTlgSOpFV1qYwThDCfTUpz/6oj2Q8ztIYwywk8l62x2SAIYdP0HYrSoJPHl2r81FNLXNsKGWUFSVYySUuSPOF33trl+u6EnXFCkhfc3A2ZJPk0KsKiYkt82+J8q4KwJBenY/jLCxXGUc7OJEEg5llShy/oSQ5LDY+kUJxpeLQqHmGW0wocXFuai3rHxFNIYUb287DjaLohHMSstSs8vlybr8m+5ETzvvsAxENCm0/qhBy/Diw3vEM5iTFZYcb2h0NHD5eUki88vcRnkg51354Tph90fXk/7+1HpU7Bzodc30nr8PCOY5yYgM5JWs5bs/nUeCzJzRL/yQttLnQq3N6P2B4nJJniG3d6TJISrTWeZbFQd7nTizjbqtCpOkdIx2DCMz17ZsEOhdLYEuQhoPPO1pib+yE7w4QkN86rUVZScS3+1JUlJmmOvyO52YsYJTl3+jECwTfuDhBpwsov/W9p+jarnRp34jHDGKJM0ak4tAOb3TCnrGmGSclCxaZTdVEIdsYxaa4plaIROCxWHW71YvpRDtp8xu1RDtLwblxbUPMd4kLz79/cpTdJqbg2WsPytJXfqLrsDDMsIFOaeiAJHIukVGRKY2lNPbDwbYtUCVwUN6ZWygoDgLp1B8+WDKKCKDOXQY1RdAlh/u6eZbbx1lR59aBRyPutU6BzUA8DOg7T1Prv4PlLTNew7kGYmr+9w1Q1eOT1zeAzLRQ1XxJ4Lp4sUCh6E0U4vWPFhqWWz889scDvX+9zp28k0hKwLLObb1YchnFMmBaUSrM7yqledKi6NncGMZ4l6FRswsy4pmdFySjJ2Q8LFJqlps+zXZ+N/9VfpFCKpU4NvzRuzFrBlcUaYVIwTAryUpOUCk+aSIYoK1mo+Yhpl6pb83hipc5q0+ednZC81CzXPW7shtzcj8jKkgudKp4jGScFd/oRri1xLWNl4ciCOFe0qw71wEVLwebQAJrlhsf1nZCtYUw/KnhtfUiSKwZRzkrTP/GC/okLHZYaEWWpzbg9Na7I7+6EXFmqcqcXn9jxyArjxhxlBXf7Eec7wRHZ90PVTp1gDpgf5IZ/0qjq+HPWfXv+cytwDrk6P3hDLKWcj64e9l4f9N5+1OsU7HzI9UFbh4fVBDMr9CRTtCs2FadmWrOHOjuHPR6urtRZafp8406Pb90dcKsXcXmhymOLVVZbPkqLOWnvMNABs1PYmaQUpUZrsyq3Kg56amiY5CWv3OlztxcBioWayzjJ2JskfHtzxM93A861a4zjlNt9w6m5vR+BENzrx/TClJ3/49/mT19dYG17zKu7EU3fNg6xUuJYkrVOlW7FodvwWai6nO9WWa65XN8L2R2nhpSdZry7GyF0iRSwWPfRWqBFRjtw8CzNKNVYliBOc97djwnTkpZvsdqs0AosmjWPUVzgeSbfRysYxgVPLlfJC8WtXkKUKZJcUfcV57tVhC4p99L53ysvoenaPHeuyR/f6BFP8800hpcxiBVaKMqpKufwCGQGe09HUB9+zewD3m8dV2UlGhoS3ECyHytiIE7Mdzm7b5gohF8ipCDOFa2KhW1J7vajOdAB09nJc8VXbw/YGKWUqqThWziWxLWlCaucihQEmjNNn7V2gJRmIxInBdcHIUWh2BinFIXxq/n4WgO0ou5ZXF6ocvFMnTf/7v+Ts02PXyw068OYnWHC5iihVfO41A1YH8S8vTmm0IJYKXrjlLNNH9eSrLarxJkJi3OkYGMQc7kbsNz0eWd7whsbY97ZnvDYYpWab9Gt+XhORs0z/lqfv+wQpzn/8hvrDOOMc02fn7iywNs7E3qxGY+FecnNvZDAtZjERn5uS0kxHdEfDvB0LXFknfvmvQH7k4zNkSE9S3m0c3K4O3O+E0wDPyO+cWfAJDHr02El68PW7OMA6vDvHkZZOOk5j5sYCvH+FU8/6qOp91OnYOd7UO9nBzD73eykaQUO/ShjlBQEtpx3Yg4bal0oDpRbs+ereRb9sOC1jRFJXtLwHX788SWurtRxLHGEfHe4ZgtLkhX0o4J+lBEmBY4ludCtca7lkReK/TAlLRQN3+VSt8Lb2yG39yO+9NoW7cDm1Xsj1vsRnarLx9aanGtV2a67xHmJ75i4ip1RjG3ZdGua9WHK1ihhlBSc7wQs1APONj26deMuen03Ii405zpVOhWHf/bVW7xye8A4KfAsgWNJLnardGs+tcBhpe7xyt0hqJKbexP6YUFaQiALNkcxuxNJbZxS8VxWpkTu3jgjR7MXFtiWRkgzFrMF9CPjJosQR8BJVsK9YUw/yRhEJeGh1kEJjPIp54P7QY0vDdcnOTXY+dDrvYDOg0ZcvjACxkOCKhPXoNURlZbEkNMTZcaSeVlQc0FIST8sqbriPt+dpIT9qZngZHrcrLVdnjvXYnucYVuS9X5syNGeRbvmstqucHGhSpKVxGXJ5iglSUuGcY4Gbu+H7E8Swqxkoebyyp0eYabo1Fw2hvC5S23OdytsDmNevTtEKWVGPLlCWJKqFNQ8m+t7IeOk4JkzDX72KTOGjtKc//t/eodX7wwQEi60K9Q8i+1xRpgrhBT81JPLPHO2Mee+pIXi1t6Er14fcqsXgxA4js3HL7RwXZthlE3jbQqqrs0oyWhWXBQZVc9hqWECPw8HeL67E86BRCNwWGkEgCGFz/g5hzsnx/k6y3WfC50K64OEKCu4N4iNf9B7mPq9FyXhvX5//Dpw+OfvBLCcdnEerU7BzodcH8Tv4PBJo7WmXXFZ78eMMYnezYqL7xyg/+MeDVeWqry5OeFOL0QKSSMwBl2XF40i6t2dcN4NOv7ah/OzvnJjn7RUbI8SemE6XRSN8Z4EUJphnLFYDwh8G5mW3O0nvLNTsNGPkcIoje71Ezzb5mPn23z8XIu9XQN0wrQAoaj6Dqofk+SauqewhcUrd/vc2vf4/BXBxtBmcxgxjHOubY1J0pSbexGTxDjFCqHJS80wyVmoGeXGYs3lbj9mvR8ySEqKqcNtXECqS3KVE0TQrOYsVnw+db7FKM55+faQfpJTlgqJNs7LpbnA7UYaWxhFmpaGnJxqyFIYpyUPWqJywD/h9khNs42O1ex5Zgnq77dcTCfjtGP06DUDOsdB6YxDY0/HljUXWoFDUeRHHJQlzCNYFOa79RQ0fJtCK4SwCKYEq1xDzTbPraVAaU1amsXYEoJf/rGLvHJnxFdv7GO0TYJW4PDxc01+9ulFmhUTHfPCmSa3dkOGcY60DSk6ywrW44ISiHPNxW5BUZQMd3rsK8jynKrvslT3+OLTS1zfGfOvvrHJrd6EhbrHs2eaLDd93tqeMEpyvrU5Yu3aHmvdOt2K5PruhM1RQq40cVayWHUIc0OCrroWqy1vvhETQiBQvHSrz2vrY6KsJHAtWoFDw3dYrvvsjBI8S1A6Eh3YrDZ9zncDtkYpdd+h4tlk03F9P87uy/w73iE5/NqHAcRsTR1EGWFacK7lca4d8Mrt/n3dnQetz+9FSfhOPGlOAcuHX6dg50Osk04agFGcszmM6EcGzBzfARw/ac61fZNzk5XsTFIuFNUTdxTDOCfNCjxbs9Gf4FiClYZL4NrUPIs/vN7DkoLdsVlIgBPnw0KY0L9nzzTIpw6lRamp2JKtYUTgGOfV7WFKWip+9slFPr7W4G7fBIm+s5NTaoUlJQLYHMRIIXEdi092HZ6/eBmA//Z//ApnmzVqrmAQ5VSSnKeWGyihzC4uTHhtfcS314dUHJv9MOWt7TFRUlAWJVXXQivFUtNnpeFSCVyUVpSlphembAwi9icp5dTNWWpwbIktjbKjn0Fc5BQlVAKHzz/W5tZezPX9kFxB07c5X5ekpWBjmBKXRiruaNPtkTaIQ52cWafGtQ8UWLOSEiqYi+DxOn6BPSxVn41H5PT2ZkUSZ4roIeSTkzxgTuugHkZUtjHg1MIAxgxzZ0fAog9PnmmileLeECoyJ1Pme5rd73ClCiquRaJMl6MZCCylsV2jtiqyqdpPHfgzDZKSb90bEueKxxYqDMIUy1ZcXgg4167yu2/vkZea51frxtBSa6quRU1Y1DybJC/IRoaD1q5YdGouIhzzl3/hRQD+m//hy3z2+fOs92PaFYdv3Rvw6vqA3XHKKCn5qccX+cSFJm9uDNkbp/TGCX9yq8cb2xN+/LEOFzsVtgYpcV6YTYaChmehEDQDm6/d7OPYFmudCo8v1YxhodLUAtN5fXKlxk89sci9YWrSzxVUPZvrexGWJViu+3z6Yod7g/Q+p+OlmhFgLNW8I0DiQflNhwGEZ0sWay4v3+qZnD4NP/54l3t9/7519UEdmvcaGZ2OlD7adQp2PsQ6ftKstX3e3TVt5lu7MVFRUHHquNbDTxqAtXaFrdHBzPpwebZkqe5xdz/kqxtD/vWrRoq9WPd4Ya1N3bMZpzmvbwxZbfmASUZ+0O5DKcXvvLXH3X5EzbV4cqXGKFG8tj7g+l5kuiixmeFvDhP+5Fafs+0AVSp8e8q7aQWcbQfUXZuq75JrTTNw+KN39/nZ6etcaPtkvsswykjykizTrA/G3BlYVByLZqvCzd0xWkhWmz5aQMWx6IcZgefgOpKnV+p8+lKLzUHG9b0Q33LZGESkpWZ7nJJkBXlpiMSztGUpzDZdYnbZw0QzTkre3g5xXAPQHGFGEaPUIp8+rlDTnT7mX9UR+JYmLaDhwmqniufY9Ccx4/Qo5PAd8xzRMfNAVUwVIOrkTozCvE9XwoWukcv+ya0BUfFoDJTZhf39KJF+2GtmJ/Cgv/dJRGYJCMvind2IMM0R2gTBNhyIs/sjQhyMoqrqChxteChRBhVfkmmBFuA7mmbVZm+Sme9ImHP/5m7EJFe0KjYV16bUpQmqLXK+frvP/iTjtbs9k0kVZ5Sl4nyrgm1LKp6FZ1nUA5fVls9+mBPuH1JADuK5XD7MCvYjkx9nSxNBsT1O+Q+v7yKknOfnvb45NqIF4C9+8hyBZ3N7L2S54aE1VH2btDBr1Eu3B7iWkacnmYml2RnFdHyHbiCoujZxrgnDmLhQOJag4to0A4dienLdHSQM4px2xeXKUnXebZm3PI9hiEex9xBCsNLw50KL6/shn3msw1q7Mn/cbC18WIfmvTowpx2aj26dgp0PsQ6fNEt1j3d3Qr58fW+asq1Zqh/YkXvi6Ljr+ElzebHCJCnohUdn1mBOsLV2wLfXJbvjhJ1JhiMl3ZpH3bNYbvq89saQb2+MuLMf8sVnV/jcpTZaa3ZGMQs1D8s6eK2ZgdUoMWRby7JB5GyNUnqTzGRS+TbDOMezJG9vjfidN3fIS0WuFCvNgLrvcHmhzkozwHMEDd9jrenxpa/25mCnF6Z89tIyv/X2DkkJ4zwnHgps4NJCg1ZgyJrDOOfxxRrSklQci8eXanzjTp/dMOXaTsgoLUnzgmFcgoipODYV13iQCEtQ9QUlxkRtEk+7PBjFi28JWhXTN9kaxsRJhu9MJeeWRVpCxZEk0vRpZsZxZmyh0Ro8C6RtsVj3sC3JO9vjI10dBzjfqVL3JF5vwp3RwZXRlWA5UOSG0HxSaSBWcG+YkheK4iFAx+VoZ2fWofhOgM5HGSi9H+n9NAWEgqnbMfePCR80NvQlpEVJnJckpXldS4KwBa6liad2AzZwoWMRFYK8KBjEJb4DsdIIBEmpqbgWWaqwBJxr+Yb/UxZ4juRS12eYZEwSze7IjGG3xzkbA0GU5WwMMyZpTm9isTFMCFOj3tq2YhQWtgUX2gFXVxqUwI3diNHo4IiwLEG35vN4zeWlW30WKy5NzyYvSqq+TT9KudfPqPkWVxZrrA8iRlFGYFv0wpSz7Qp/6soiizWfzVFCw7NYbhoDUwHUvZRCQV4qvvLuPl+9sYdGcLbl0fBs3toesxumPLnSpFVxeHypxlrLZ7UZzEfrvSilF2ZkheLxJSPGSAvFztiEje6MUy50D7rbs+DO/Ul2Yqd8Vo3A4cmVxtyFuBE4NCvuibELxzebSV7eNyY7qd6vgewPSv0wfK5TsPMh1uGTRmvNS7d6OFIaiWa3OifXHSfPncTyf2NzzB/f2seRcu4G6jvGu2YU52wMIu7sRxQKmp5N1bOxpDEeO9cyjBFLmhworeHazoTf/Po6O2HC5y50+S9/7MIc8BwYWMFjnQqfudRmYxizNYhIipJ+LHh8scrji1Wu70XcGaRsDWOyosS2LJqBy4trbf7z51doVFy+eW/EKEm5vq84RDXixn5Ee3uM0IJO4FCWxlK+KBS7k4SLeZVLCwFJ7oPUuFLzifNdnloJGCY5oztGMrszzkmKkrxUeJZFlGY0A0m7YuNlkoolGWeKUalISo1tGQ+cKws1Fio2UVayNc5QqmRnnKOU6cR0A0GqLNoVi70wp5gaw7kCCgFJYThAChhkJdnNHitNj1IdRS3twOTz7E9Shskx2CCYprffX/b0taKp6dwog0mW0/Z4oIb6eBfnpLuddJF/WH1Ugc6CA64n2J+YHLX3qsOf2eLkz/WgzxqWGJPAaeUYl+0s0XPzSVtCxTVeNbd6CZMMwjRluekTxSWuDVeW6pxtBryxEyLKgmbgkinwvQJHSp471yTM4MbeCK01+1FOVCjyEt7dDVmo+qAVnm2jlaYszXfcT4zDpRTgO5JPeDZ132YU5wzjg7Xk8aUaUV7gpxbDOGOUlEjLvGfXgr1JziQtcGzBUys1bu5PcC1JXCoqjsUgLjjXrrAxSNgap1R9hzNNn+VGwO4k5amVBpYliNKCl2712BylNH2bTrVBJ7C5tttne5RiS8l/9fmL1HzbrA9xRrvi8uRKld95O+a19REN3+ZcK+DqSh2t9dwR/ni3xbUESabYHCVUHDPa1vr+i7KUkp95apEXJg3qnn3ipnJWh8OUHxY9ceTYmdpy3BvE8/f9gwoMDtcPS/L5Kdj5kOvwSTNTDXQqHk+dqZEr7stF0Tpmue7iOtZ8TpwW6iAMVCk6FTOznuWlvLU9IpsShx9frHFxsULDtQkLNV38LJ4522ScFlMSYcDtXsS7exPirOS1rRH7YcZi3Z+j9y88vcSn4zabw5hX7g4JUxOP0Km6dCoOnVqAUiV1P2UQCwLXxpaCMw0Px5JsjhO+dmvAmZbPH17bZZIW1DybJ1fq87+NUtCLci4vVXnmXIPbexG39iM2+8YHqOJJ6r7Dzjjkj2/sIYRkpeGxMWhxtu5RnK0zSQqSoqQ3yehFirwsUaVia5RPc3c8OhWbbq64lo+ZTZe6dZ/PX+qwMUh5Z2dAlBekRUk2vXgkGUT9gnZFE1jmwpGWBtkkJWCBOBbguZ9CvJdS8SSBZS5QpoMkubk95OYgn6tuZlVgxmNgSKtKmS6OjTGr81ybdFwccXCeHLqy2xwFNA8DMRamI+E5RjH0USAwy+m/D+J/M8ohwJg1OpiR0jB/7+cKMGA2zh/+9zrMpXrQcx5OPVcKklzzzl5CXijKwnC1QBN4Fq5l8eRKnWdXO4T5Nnd6IXcGiRmxIkFKNocZFzsB3ZrHvd4E17LwLEPCX6n7LDd9Lna7lGXJJM8JwowkV9Q8m3FSYFmCvFC8vjlitR0QZQVCHxU/3OnFbPRjLCGI8gLPttgep3RqPhXX4vKSy/lOhSdW6vy7N3bwPZuWLbm6XGMvzHn+bIOzLZ+7/YgbuyFnmwGDOCOd5uc9uVLnW3cH1FyblYbHWrvCzz29zCQp+ZNbfeJMMUpykrzgD9/d5/ev7aKV5tJi1fzdhQEweam5N4jJlWYQ5yzVPD51oX2fw3BWanxXcqbhsx+mfOVGb24WeHzT+O5OyCt3+gC8eKHN1eX6ffc5KXrhsLT98GbzcM1sOdYHCTsjw/UJ3kPh9UHqe91l+WFJPj8FOx9iHT8oj5PXZpOjwz4ScVbypdc2EULw4vk2V1fq8/wWgHbF5emz5gQdxTnv7k0YRDlKa9baAeOkQCnB9jjj7iCmGTistSv8zFOLdGsugygzUs7FKlcWamxPYp5arNEO7BNtx79+a8C9QURWKLoVm8C1sKUkL0oGcYZGIIUJGFys+lxZrvCtjTGOlLy9M56G98Xc60dGAnrIHLRbc4lzhdaCumcziI3c9mynwqWFgCeW69ztRbx0q8fuOKUsNPf6IW9tjWj6NrnWnGn4nKm6bI9T8rLElgItBUmqkaJkGKUUhQVKmfFBUdCquCzVPF7bHDGKc8Iso9CSPDt68cs1hFnJrX5Cw5c40mISl0TTiAiJGRkdBg2xgrrQOBbE0yfbnJRsTk6+rEZTlZjNNF/LhnSaqD1IQafFfaCkeMD/w8MBzCygtMy/N92a6hQQTt4DVX1Q5X2OySmbPUepDeh5EDCZja4aHihbomeZIA+ois194PSk53SAmmc6PQpj8qkUaGFCXi92K9ztZygUw7jkUxebaBRfeq1kc5jSDmwqDiZeYXtML0wYRCW2lDR9Sbfu4tmSn7y6yPNrHQJH8JXr+zy1ZDoerao7l4vf3jM5W2FacGsvoh9llKMD4fz13ZCJNQQBz682+PTFBTYGY67tWPg2nJtK21eaARXP4emVBp4jqTk2rcAjsCW2NJlVpTZjsSgruFypUaiM5YbPQs1jrVvlhfNtniqavHCuwXLTpxeNuLRQYy/MONvw+cbdIS/f7ZNkJYOk4HynQi80zvAXugb4LNd9+lE2NdyDCwvVE+0yVhqBMUXMzahxe3yyNPze4CBMdGn6Og9LLD9Ym/URaftJXMdDkOnIz+9lM/J+gMv3o8vyw5J8fgp2PqR60EH5oJbp48s1lhseL90yxEMQLDViLixUH2j/fXjcdLlb5dMXW7x8d8goztkLUzO2KjXrw4QzrYC4UKQl7E4yPnWhza/9mat8c31IVmje2g4PLSom++XazoQb+yFJVrBYc6n5Lq5tcWN3wvXdyXQcV+Fj51o8d67FpcUa59sBzWCfa7vjqXqipOnbbFnwxuaQ60nEfz393C+ca6GrVQZhxusbGXuTjIbvstz2ubrSYLHu87WbfcK0JMkVvg1ppkiLjHFaYEtJLywoSrNTVEqzVLWwbJsoV+SlolO1iDLjmKoKRYlgmBRYg5BMaYaJwhKauidwhcDNNVFhlqvZV9VPSpKpp5CwFCrXlMqMLWZkZRdz8ZXAONEnqq4eeKxMH5sVHAmEfBAIeNDts9HMw15acZDu/qggY7ZIvN/uS1oaovfD6jvpLh3HKYPiZIn/4fsXwDAFchM98qDyMN/pw0pg4h6WGh7disMwLdkbp0wSRTr9YFoLPrHWoFmNeXsr5PXNMf+vr96hW3EZJTm2hE7VxrMs7g2SKS+lJFWammORloqGLbi6WOdsu8qVxSr/329t8pUbfSZJzsVuwH/21AqXlmq8vjHktY0xW/0Jm6OMcZzdZ4cwjBOsMOVsy0chaFUs9scOjiVpVjyuLtdoBD6DOOduP+HFi22WWwFnGi7Xd0P+6LVN/vBdh5+8usCFTsDOeOa2rud/ZCHEPHomzgr+5FaP37+2R8Nz+PzlDr7tEKY5cV5QdWyaixYV16HqWeyNDWD6c8+vkCuzxl3fjQAjznCmcQ7HwcFaJ2Ct7XOnH5sonam3zuH7erbkXCtgZwr+zrWDIxfu94peOM7ZOQ5UPMfixQttlho+59oBri2Js2Ie0HwcnBwekXUqHk+frc9jIB5Uj9pl+W52f35YVGanYOdDqg9yUDaCgxA8gHOtg5PxJKB0OC+l5lnkpTbqAjuh6tnsT1KEEJxrBfeZbHmORSoEpRaEWYmMMjoVj1miugb6UcZi1SEPbH7s8gKDOGd/nHBnEBOlBY608B2b59fa/Omri/iumYP/zFOLRHnJej/EtSQvrDXZn6Rc255AXPIHz/8Eq62ARtXlVphyey9maxzjSbNrHMY5N/dCJBrPkfiuxXk3oOZK+kmBYwnagYvnWtzYDXEdgVKSiutwvu2SlGaRqziCXGnCtCArDUEUoRHSItaQ5UambtvQqXokecEwzvE9wbmmx05ozBzjFLTSaHKkkGhdGuKvOgAMgQNWYfg1xSNcwU8CG8l3cOWXQN01uV2P4rHzfropBQeBl++nCo46Rn+3ysck2YfF/QAs54CzJAHfNiDy8P0SOPIHOsxxcoGqa7ps0THmc12C78LuIYfBemDRqXoUWuCIEs+1ifJs/vxCWry1EzKKC8LUGGqu902QpuHTKFRZ8vy5FrYEpTVRofEE9KKEcVTSc3LysmSl6bNQ9dBao5RimBTc2Iv50uvb/JLnsFiv8NxZ2B+nbI1GVF2LS92AfVHye8/9hOG9NCssNX0eX6oTOBbjpGR9lBiVkhDUfJdJljOKC/Iy5nOPdbjQrTKKM/7TGztsDBP2w5QnVup8/Hyb3UnGUs2jH5tx2iyPr+7b3O0nXN8d8x9f3ybKFWebPj/91AJRpri5P2FnnHKm6fPESp2nVmp89UaPazshX762y91ejXpgs9Iw/mBZodgPE377zeQIbwY4uqlcqnGhW8WR8PrGiH6UHeHYXF2pc75bQcB8HHbYrf69ohf8KRZ50Gb26nKdC53K/Pf3BjE7o4R6cL/Vx4xY/fbWhFyN0GieXW0+FEw8SpdFa80722MTBt0O7hvVfZD6YVCZnYKdD6ke9aA8fsLMTxaYc3YeVlIakDRP8617fOpix3CB8tK4HAcOUsr70Pnh8dn/n73/DpItu+87wc+53qSvqizz3qtn23cDaKABAqADKYqUKGlkONJoNLOSYiWNYmg2JEr6Q9qYYGgnVoxRxIwmZkgxNLsSpdUOR0uFKFICR6ABCcKbbgDtzbNVr3xV+rz+3nP2j5PlnutuAiCaZP8iKl68rJM3b2blOed3fr+vafsOF+d9rhgh7gz4PEm0X87lhRrvO9fkxkGCUopL7ZCb/YhKaqq5ZYgjsJ9Sil6Us9aLuHUw1a/brfO+c01uD1Oul4q/9+f/Aa3A4dK1EVe6IaWqCByLrNCNlr1RxqvbE273As63fa7MBfiuyZmmxzgtSUvJ9z3cRamKf/e1bfbGCXXXopIwyRSNmcBZkpdItCJsJ9BGh4O4xDAMLrRddsYpo7TCNg3aoc3+RNIJdaJnOha1QjKNyyNPpFAoWoFDzTHI8oKD5PjvEBX3TwhcQ1PTA9dikJQkud5gD2nOh3GygnK4WR/iRt4seXEMTa2/19i6pcG1d9Kj3058M0WevxHzUn+mYxS4BvlU3pXsHN6nAQQ2ND2TrKwYJqcTnpN4nMPWlwHUfUHdtZnm5czs8jgcGzxL4KDImVWKKhimJcM4R0rNspwPHSQKW4AwDV7cGCNMg05gUHMsXFswyQryUqFURVIKelFB4Ds85FlsDVImaUk/rkgKsKuKshfx6Tf22BylvO9siw+eb/PixpC0UhSl5PmN0cwKBsZJjmmYpCW851ybuNvgn/7oP2IaZ8zZLp3Q4ZnzHRxL8IUbfaZZgW0YrLY9Hl+uc3uQsjkYMEkL1vsxQgh2RgkNz2a54RK6FhfmAh5ZapxyCQcN4n1+Y0g7cBjEOUleoRBYQlBUkpe3JoySEs8xMQwtgZGUEsMwSEvJjV5E3TWBKSstbQy62NDJ1CAuj3wBARYbGrt4L1uIlzfHfObaAbah18+TWjmHWjpwbzfyt1LBSIuK9b5Wmb6TAbbeT7g9iNkbp9RcPat9+26pD9cy6AQuhRxjG4JBnJOV8k2p7fdiip2836yo+OragM1ZonW+8+bq0H8Y4t1P4FsUb6X0d1f1p+PDrBz6dnq7p68D5zv6JLTeT+5SSr5TrvzyQsAkKXhxY8hLWyPev9rmocUar2xPeG59QJSVLNRdSqklzc91fBZqDr/43AaDac6tfkI79FjrxVzphryyNeGFjQE7I126bfoOm8OYx5Zb+Lbg117Z48Z+TC4lk6xEKVioedw4iBAYuJbJNCspyoqvb4x5Y3dCO3BpeAZ5IRlnOoH5+vqASaFND/MK4izHsU2EMCgriQS9gTgmjcDhfNtjqRngGjApJCsNh5u9hFe3x6S5ZK2XkhWa/uvZJmVekktFrnQS45jgWia+Y9GoOxSly2BzelS5KNAbrTv7bBs+R35YBQLPs2gHFpO0QqKO2gvOTOywFcC00CBXo4IM/VwBBIYWI7Q5vN5xWGh9FoVOuIx78MSj8rjddvicb5eBqAG4QrfrfjeFrERBWoBCYt3hQG6hgcdlBbah8SQVAseyCN2SONOfQacmyCvICkU+46GL2d/Zsw1Cx8Q0FWkhqU6AmHsZ9LLjD7dmQcN38W2LxKqIsoKaY3K27XF+LqA/zXlhe8okKTANQdP1cVybopQ6Oa1KklwfKPamKWeaPmHgYpsmL2yMMISBa0mdYNkWO5MMx44JLINRVtEKXZq+xZl2gG0IbhxEbA0Tskqx3NTA4E5gMc5KHu6GXNuRRCX0pgUH04x24GpQszB0klbz8ByL1Y7PzV5EUckZvkXjkC4uhDy61GCa6wrnYQghuNINaQUWL28O2RmlSCmZC7Xp5g891qVUinFScJg/NjybRxftIwwMQlD3LC7NBZRScqYV0g6tU9YPSikC28S1xVFStVj37mJpZTNbCdswKKSkHThHicGdVY/VTvC2wbdKKdZ6ES9tjhilJR++0DnSPjtci5NCf2sC19IH2LngrnVdCMHjK3UU6qgC9VbwMHcyxe6sLh1/Q2ftsje94h+OeDfZ+RbGm5X+7tThWevF97RxOJqg/ZiFuseTZxqnerv3us5R+fQBSslKKV7ZmvA7b+yxN0lpBQ4LdVe7DI/1hI1zDfhTSh29n9VOwOX5GpuWFhisuSa3BzGTJOfTVw9mCs2W1pbxTHbGGXk15I29CVkpCVyTucDmybMNnlntkJYFlVIMopyikszXbQaJPhWCQTZJGGc2V/djXMtkrmbz5fUhDVdr/ZhCELq2pt37FitNl+v7E/JKkicSUwhGrslCQ2FYNis13fsXRoplWoyzWGsEFSVC2Aih0aWlVISOwDIUhtCMmCwvKG0X01A0PMFBoo4WEwW4DjQ8kytzPtvTkv1pjsol47hAVRLDOHY+Nzlu8xQVuJZAVYroRBlFMLMVQCcHtgmyOl1pkUqPMdDXLu5Y3U4mFVoMUSdW98Lg3MvD6+Tv3i5z6hDPdEggk8D0jvt7u/YWh4ldaGlMkGtozSQTsEywXVMDxPOKybREGjrxPGRt1V2XJ8826U0yXt+dchCVR2rVhhAIw6DuOGQlGElBlN9d2TKBM22P//KZFcZpyRdvDjmITELbwHdsemnF2jClqBRKCGqehWGalJVkmEmklCAEwoCyqrg9yFhq+lzuBOxNc7aCmLpncrbps9T06EcZ+5FOjG4NErJKkheSdljnoW6dmmfxO28ckJeKi/MBq52AcVzwC1/ZYJiUtDwLJXSiGbjmkdO2YxoIQ+i56ltkpWStr6sSliF4aKGGMAR7k0w/J8qIct1+ySupWVJ1FxRsj2K+cnPIfpTx+FLjKAk50wm4shBye5CyN82ONv87PfqWm9rOZpKW1FyTuVBXWgzD4Eo3ZLHhUnNNplnF8xvDmXVExjPn25yfC09VrE8yXw8JHXB31WO17b9t8G1W6iSwkArXFJgzLI9nnFyLNc1/tRM8sEJvGAZPnmn+rvA194NKeLbJ+1fbdBuaAv/7vf30zYp3k51vY9ypw/Ps2uCeJ4zDCfrixgjP0UqmT55t6t/NJsmd1zk8WdyplHyyQpTkJV9dH7A3SdmfZNiWQVUpQsdgoe6yUHOIZqDLWwcRjyw3uLo75dlbPW4cTDEMwflOwDjJ6UcZ46zkYJKTlBVXGjWeXvXYHqXc6ukT53o/Ydwf8Ov/3Z8C4H//ta/zlbU+oa0VZjeHKUpJGr5FzbFouHqBDhyT3XFKUQmUKqkqC98VLNQsPCfEwKBCIiSkpeS1vSnpLAlLC0k/LRjuVviezfc91KDbcmc4C9gbpQwTk1FUUSqIsgLHNIkqA9Mw6M5Ou9MsJ81ztkcZtwcZtqEnz5wH42ym0CzBMEzSSrA5rTCVREkNhB3nEOcVFTohcSwoy+MNfphpF6TD/x8ueTVHj5WaQKKTFVdQlgrX0Jv+YZUmdPVia2bylCHpSXq6xioIUqlO6cYcxoNOgYeifHc+9maA6YoHt64OJe8EutLlWDrxiGasMc/Uhqsnn28a0PQNokIiETiGwTSvmOYgqDBOJDii0gmVLfRnd2WxxoW2x+4woZLHd1+g/4ZSVlhSkBcVtmlimxXVHQlo3dakgCgreGFzyijNMVBIdAKQTxUGgpW6S1xWXJ7zkIbB/qRASUmcFYziQqtzC2j5gqqs2JvkXN3XraPHF5v8+WfOcKUb8puv7PFrr+whhAIU00QDfHeGFs+uDfjeh+f56KUOb+xNAe1O/tXbQ67vxzSqlP/jb/9pAP7ev/gMT600sEzBwSSn6Vt87KEF6r6lkw2ltOrxJKNb944MMg9xMJ98NWVrlGAbAZAxzbS1C0A/KtiLMlzLZJwVrPUTNgYpb+xNEULwxEqD8/Ph0Vp0svW+2gm40g3p1h2eWxswzSVmkh8Jrl7bi061mpYaPocu4XcmEw+qqt9Z9eB3Ab51LYOz7WOg87m2fu7h2vpWW2En7/d3k5DcDypxiE06/Kx/vwKKv9nxbrLzbY6TJcn7nTAUmlU1zbWz8e4k5UpR43Y/OVXCPH2du08WJ8ue3ZpLVlas9WOkhOWmx0MLIUIoXt2esjcT6HINrXXxlbUByy2fjWHC7YF2J1+c9c6v7SfkZUkhFYt1l5awaAUWyw19siuKig2Z0HBN9k/IBP//nt2kLxyavkXTtVBKEeUVlRJaH8c3kZh0a4JRXJJVBYFt89SZBlEuaQYOi02LmmuwNkgxleRLa0PGUUpRKgLHwDYFUaaVoLNKsTVJODcf8thSjY1+hGVZLAQ2SV4xnHHFzaTCsCCwBb5rs9TweH6z4GCqjujkh2DiQELLA9+1KKQir/TnPIpyHEthmwallFTyeLNWCuzqdMvpXpUDz4LlpkPoOdweZMR5odtApsDzDJDVUUJwWHUJbBtBgawkqYLGDK9zWEEKTGh5Nlmen3LyPvldu+f3dPZzL8XhO7V+Tv6O2X3Zs8rLgxzeLWC5YXFp3icuFBvDFNs0ubLg88LmhL2oOrq/mi2oeTa+p80zJYJsmHDkxiFPuws4NrR9k8uLdVxD8NmbQ3qTjLnQwRAZhtLUcceGYVyyVVbkpaLtmbQbNgdRwTTXn4+JdiHfmmRcezGmF+XYhiQvFKZh4rgVTyzXmQ8cJnnJ9b0Jo6TCtSTjmR6NKQS2ZVC3tZZULuEgyvHshKIqqZQu1b28PeHlnSlfXeuzM9Yec+c6PosNj/1JxvY4x7FiPvnqHpe7IVfmQ4Qh+Nz1HgBzNZfaCcR8K7DoRTmbVw/wHYuL8yHfeXkOYRhHOD9gxq6aUahna1RaVEd6NjVPV14MI2c+tIkziYHkyaUmUVFwfi5kpeFxsxdjCdgdp1zp1o50Z9IZPXyUFGz0Y91WavkoFHuTDKUU5zvBkXP6nRWMw0rPYVXoME4e5u6VQHwzqh6ngMizawJvmRJ+P3Poe1V3HgRfeFBS9wcBUPzNjneTnXdIPOiL69kmH7rYBhSWITjX1myCB5rVdfxjOugsjgxD44JpqgEi86FN07PwbYObvYTdSc7eOCXOJZvDhElWUXNNDBQoxZmmx27Hx7MN2oGt8TFSkpYV59oBl+cDnlsf8dtvHPDV9RFXFgImecmFTshK08HKjlG9mayIZYVjGXRrFp4p6NQEaVnhWQKJQZQVPNvPOZimpJXCRAsRBq7Fy1tjKiUx0EySsiwYxhm9SYZjmzimRVpKlIKmb+IZMJimfOnmASaK9X6MlJK4kCR5qZVwZ0BfQ2mxspWmy0LDwdvVekJ3tnkKCfN1nwvzId2axe9c7bM/2xWzCgJL4NsmSVkdZRIC7Zp+Z2XFBDwBoacrGUkBu+OC9zQD/ux7Wzy7NuR2P2KcS6ICshPCgBJdYQqdkrqj2yUqh7g8XRERArLqwUai9/x+oltFbsUptWIPjY+ZPKAHdVj1erMDZgFsjUr2JhM8x6QTWlyYD/ngapP1QcooqbRBqwUrDYtpoSnegQHn5gN8Q3JtPyOuToO/PQPqjmC5FTLnO3xxfcgkKfAsg6Wmx3nfZpxk7E1LRpHEMAVlpdu2rmvzxx5fYG2Q8LX1Ib2oxACank3TtxnGJShFnCtKCYascG1o+zaPrzT4zLUDCinYiXKqomQ/0roGc6GJbRmIGaPRrBSDuGSSTVkIbK4shCyEDrf6CWleMk41sL7lW5xp+WyPtBihaULomry8PebrGyNagc17zjTxLUG34fP+cy0u+ccHp/1JQWzHhI6FZRosNrwjFiVohtLTqy2KSmGbgvVBwsOL+veOKWgHDkopOqEWRs1Lye+8vs8bexMuzof8xQ8u86VbQ/JS4tgG33mpw8tbY/YnGev9+IgZdFiVyEvJJClI8oqNoV4baq7FIC7YHae4lqkrPjWXrKhoeboFfbLSc4RVeQtU7jurHvDWk5Q7r3MS9HuYvL0V9u29zKHvib15i55fJ3WC3m7C9Icp3k123kHxIB2eR5YaXJgLT50kTuJ0tE+TOvoyr/VjvrqmlUKfPtfiwnw4Syq0Yejt3pT9aU7gmDy92iF0DPpJeVRdibKCYVIyTAqEoUXKvr4xYrnl82eePgOzifi12wNagcWVhZClpktSSuK8JMsr9gpJOzDZGKREvhYhPMQQATy61KCLzdmWx+PLLc52fKKs5IXNEQfjnH5cUFQK09Cgu7JUGMBiw2FjkPHK9oSs0GTjVugwjgst9icgLyo2hilpqTBnOIVJUrI7yWj6CVvDmP1xxs4kYzDNELMWRzd06dQdvbkLg0cWQzaGGVVVYRkzj6QTlHO9oUoeWqyzM0oppcSUuvJjGoJMgmdrQOphcnMIeJYSLHWcPJjAUtOiVIreSLe8skzx8uYYzzYwBFRKt7AON3NbHGvzVMDepMIyKs3KUndXYrIKxmn5puyqOz22JLoqM+eAK7VVhmlqxlN1B+BmztZJVgnYpqDpwihRR9ozR99rdNtKqePXytEeYWVeYQpBlpfsTHKSoqKafe65gkmuGGUVcabYUbA5GnFuzqFTt3HSkvHMxqFCCz2WkcI0I/rTjEGUo6R2Bfdtm91JwiSpkFKhhMJF4FgChMFczaHm2VyYN7i2FzPJJEpJlls+3/XQPC+FE9Z6E7bHOUlWgjBo+hauKVjrTdkexmSFroJOUoEVlwjLoOU71AMHITRTaZIpUAUGkko5dGou8w2PdqXYmWQ8VXMopRbRbIcOO+OUKK+oOQZtz6TnWGyOtCREzRF0Qp+L8z4Nz4HqOD3Nqoq6Uvi2wYcuzPHESuOuDXCp7tKtOxRSHflQuZbBtb2IfpQxTSsg4/q+yULN5ktrA7aHKf04J85KPvnaPkLA9zw0z594apmDqCApqlOeVoeg5m5dv5f9ac5C6FBUil2lmGaadbk70cQNhWJrqNvtChgmJzTBZonFW6Vyn1xn30qS8laShbcqvHevKhXc++D6dpSL7wtW/gNi9fDNiHeTnd8ncfIkcao33PFZ68U8uzY4+jJnpWRjoJWLlToWFjzXDjjX9vjM1ZIvz9yTryzU+M4rBkutkNWoYG+SETomm4OY63sTpAJLCF7dmWgzTanLy0oIXt4e89rOhEle0fAsXt+dalCva3GuE1B3LDbHCWkhuZ3FtH2bxfqxVNsPP7lCZ6HNY8s1bvZSelEKGJxp+CS5oh3aZJXEsUxcA7Zn7C5dYZcopTCEIJcwiguyQuLaglLpqkcpZ+BhU//eIGaSSnaslK1Bgu9q76C81MnBlfmQv/7RVdaHOVvjjL1xwm+/0ef2QFeADAHNwGAUS4pZWypwIHB9znZ8klzSDFyirMSutGZKWWraqW1WmDP7iLYPhqlZJa4pmUSSYaU3+u2x9jc6mYz0E8mzMzfpopRH9HLXAEvohCmaJRKHycKReusd3yOtrvvmDI38Ho8dAoN9a1YBExoXk5+oErkGXFyq04tyxmmBklAqdZSogF50DA4ZU8wcvU+/lpQwzUte2Z5wrRcTJeVRhWpaQT7WlbhDMHaq4OZBrmnplqA2s8Q4vG8ltFbSONOVFcsSXJwL2BpljJOSpNA6K4ZhMBfa2CbklcAU8LXbI9KiQioJSmIaBo4leHSpSbfh8ZVbFoohw1hbuJiGySu7U3rTnEGc0fJsvvfhOUxh8ltvHFBVFb5tMk4LhDB4bKXGx7ohX1sfcW0/Yj/Wif6FuZC5tseHL88BsDfJaAcOuyMNDt6fZkxmvm9/5NF5fvPVA/pximNZWKZgkkle2RnwkH/8udZnrK7Aswnd09RlgDd2Jzx3q8/mMOVMy+PhxfopS5t+VBzZJhhGSremK8M9S+CaJtf3J+yNEpKy4sJcgGUKunWXQZyfEvo7pKvvTlIWQoenluv64JHocWfbPvvTY82vzZngImiPrMWGf6QJdtKt/O1Sud8sSXlQsnBPhfyO/8C5db/Xu9djb0e5+H6J0R8Uq4dvRryb7HyL45tdQrxz8q12tFvwka/WrI99CKArKoklYJwW7IwTurNko6jANAyUgIbn8sRKnXMtl89e7xFlJXFRMV9zmWQlrdBmvuZyMM2ouRZrvZjVjk9ZKUZJSVJU7Ec5ZaWplk+sNPn+RxdwTMF/fGGXwLEQ6NJ+MhofvReJBsoWEjZmRqZFVbFQ97g072MKEIbBQs3mscUaX7k1oBfnvLQ95WzT5bGlOnFeMElL0kIhAkGnZrM9TJkkBVFWYRpaG2WuppOQpKrIpd7EDNOgU3PZKyW+a9EIXb6yOeWN3Yh0ho/aHCWkhcKx9AYn0Mwsy9RVibpr8/S5Oqstj1d3ptRdg0bgEqUFWS5xTMhyrUXiW7rydnYuREpFPy7JSoVta7q5RFPMBafbZRIYxpLgDj+nXM6qKxZk+WnczJ0Lro++qCmg7kI/5kgr5u1EOWvPFbOMKxWn22SVhEcXQ6I84NrelH6cUhQ6QTPQyVC3bpFVFUmhyDPt1aWX9BOvwwxrVCnsorwLGa2Ufi9w/PoSbapqGopOYOBXikGsE63Q0X/vPKuQCkSl6E0zPFtgm1ruwTYErdCl4dj4nkU5ozDvjRNGcUVWSa17I6RmVcUaMwOKs62ATqDbMUIIHMtkvReR5oqJKJHS4C9++CwfvNjhVj/m6+tDtoYJtmnQcC2u9TIKqSuUldTvL3S1sjFCsD/JSEvJG7sT4lxbSViGQSF15ee/PtfCtkxuHUQoITjbCtgYxDRcC6mOU9dJUSJcBwVsDBLyUrE9TrkwF3C+E3Bzf8qLGyOmeUEntFlt+6faTlJKbFPTxLs1F9c2+aEnFlk7iEjLipe3RiihqHk2jil4bXtKVJTMhR6XF4KjBKfl2wzinHFSstGPaQc2e5OcdmAjhOCZ820uzNeONvgzLY/NgYttCs52Ah7q1u5yIL8XlftOFeU74yR84BAfdEqz5j7Jwr30ebTv2L3Vku/1eidf516PvR3l4vslRr/XVg/v5JbZu8nOtzBO9pDbgcMTK403lQN/s7hz8p3r+EfCgGku+frtAXOhx2PLNc53Aiop+ey1HtcPIkLbouaafPB8h/5MHfhcKyCrSq7uTrnVi/ja7RGhI7g4F3Ku7XOuHXJpIeRgmrM1SvWiNM1Y7fgs1B3agU0Hi0tzIY3AxhKCM+2AYVLqSd/QJXFmJ+s3hsfJjmsZLIQOX7rZ5/PXD+hNMxbqLmv9hAsdX7O4Rhnd0OGhboAQMEpKhFIcmAYLDZedkWYV2ZbiwnzAY0tNXtgY8OLmCMcUNAKL+dBluekxSCrWDiJ6UU5WlrSFzSOLIVcWaqAUV7p1NodaKEyWFaEtWAxtcl+x0vAoleLqfkQlJEWpKzS2BZ+7ts+Xb/VJs5JpXpJLgW0JXV2qwLAkSQbC0G0d2xBYtkWpIMkFiRDYeXlsMYBOYPIZ3uaQdh7dYd4p0C072zVxyuqBys1n2hZIRS+uyJXAtxWquHcF515xyLo6dPs+jDstFxwDbuzHXO6GLDYcQtvgei9CoStjNU8rYk8mFdNsds1Ct+Os+4giGkJT+st85n9lwkrDIJUGUgoCx0TJiqyoGGUSy4BxJrGEwHe0hIBnCpSqaDgGcVaRK7i+l7DS8ah7DpWUdAKLXEI9cDSAfUZP3h0WJPL4vgylr//q1phxVtKLC9YOIhCCxbpDt+Fzcc4nzgpe351imyZr/Yhnbw2Yr/l838PzBLbJF28OqHsW86HNziSnKEoM4dD0HZ652KEd2kxzyStbY4pKY89AsVBzeP9qCxBM04LQNWavY/Hk2Rad0OXRxZBff2Uf0Zuy4h63jtu+w8TUAn5xWvDrtwesDxIuzYV8/6NddscZSSkJHRvfNo+AVodtJ828SukELgrFs+tDFmoOczWX//D8NklecKYVsNLykQg+d+OAwLF4ZEkwzfwThsdKV2EqxSQpWO8n7IxTIDjCsRwmH44pEAhWWt6RSrJhGEdqxqe+Kyeo3LYBr2xN6MfZfZ3KD9+baxn3rODcL1m407w5K3Wb7n5qyXe+3p2PPwjC8FYqMfdLjH4vrR7e6S2zd5Odb2Ec95AnFFLbFTxx5u4e+duJOyffoXfLYsPl67cHvLEbUcjJUb86K7VK6XLDw7UFr+5MGWUlP/TkMot1h1e3x+xPC+K84tpehFSStV7OubbPlcUGf+o9S5imSVZU3OpFbA5nGCFgEBc0PJvFhssjS3Xe2J2yNcrYn2iBv5pnMYkLXMdgmpTUfYuoElz74Pfg2yaPn20iPIvffHXC7V5CP85QEhabHsOkpBcVJFnBi+OYnXHCfE3rbDQ9izgveWMnpR+VuLbgkaUm3/vwPJ3QYzSjrG4NE8ZpRVzAe8+1eWSxzq++tM1vvLJLklfERUXdt/mBRxZZ6fh0Q5t/8fl1ru5HxLmin0ksQ/DoUoOPXJxjGKVsDFOKsqAEegmMswLbKPRGqPTeINBaMmWlN+ao0Cf1qgRTSK73Yq3HYpkElqQswVTH1RwDPd6ZWVCU6nTVxkYnCRKdEBlKUnMFyUzzRwChcWzA6QK9RLOLihJsFKEtKGZttkPxwvtVeUw0VimV92Zj1czj+/NswcYwY5oXWqcGSVVpAUAhoOGbDGNt33F4rXL2fgMb5gKLSkr2Y0k6Y6xZlsGlBZ/dccY0V3RrDh9YbfDqTsRBlLPc8PjQhRYv3O7ztdtjRrMSkbYb0fc/zRWeMGlYCtecJZIS8qKkXfdJcmiHLkUF7dChQldWKqWxRicTMNOEnVHMZ69XuI7NJMnYHmVIBArFH39yhe9/dAHLNFFilyQv2RtnfPzFbZJC8shinSdWGnzgfGvGODL4ytoAxxRUlaRTc3lksUaloJQZE6AV2AwjjU+LC8l3XVngmfNtPnetx/40Y32Q8vBijfeebeFaBlFWYZuwUPdo2hW73/X9jNOCuabHxWadVmCzN815bWdClEmansXWKKVTc3j/uRYLDZdL87VTG21eKQZJzjSTlFLjgNJCkhcVNw+m3B5EoODxlQaPLTdmJruSQiragXPKsqZbc1nt+FxaCLi+F/GZ6wcsNz1Wmh5FqfjKrT5pLvEc40iNOSsVw7Q40rS5XxwmKS9tjmYqynrwg0DD40RXvu/EAd0vWTi5Drd9h36czXTB7q2WfPg6h8nbnVWpb0Z8ownTNxrv9JbZu8nOtzBcS0/UQipsw6AfZw/sIZ+cDFkpT/m3HMa95MKzUlL3LDqBS1KM8CzjqF99KLAlhKDl2eyNU5JS6usshKwdpLy8O+axhRoPLYa8uqMYxiVxIdkepzOKt6Xv41CRUyrSTJfsDVNgmga9qGB9kLA5SCiqirPtENsQ7IwTtoYphZQ8udLgqUtdbv6Lf0OlFP1bE6Qc0Yty9qcZga23+2laMB86XGr7vJoV7E4qtoqEaVbygfMdbFPw+k5OLy5wLMHqXMjjy3VWOzUGSY7vWNQ8i8WmR1LEjOKMz17vsdr2mQ8d6r5NXEiyUrE7SvnVl3dYbfuUpeLq/hTXNBmqglFSYJsmw6RifZzRrbvUPS1YV87cyqXUAFhTaH8sga7gFDPRPmPWbjFtcCwD3zSRhmbFKFFgm1rozUInMaZ1mBSBZQgMF0yh2E90MmIBiw2DnbG2SsjR4F/bOqaACzSI2mbmpG6iE52ZGKE5Y5p5tkFeSYTSlPRS3ttS4hDz4hj6vd4ZDV/jpspKIZUgK0tu9ypdhTE0s8wxYL7hcaallauTsjrVqlPo9tpy3WEnKqkqXXMqFHgKokJxrumxOc1J8oybBxHjrMAxDboNh+WGyydGJfGJ5M0wNBDcM03iStP/lWHSrUMxlggDECa2AbZn0osL2oFFVlb4pslUlMR5eUp5uubActtlf1qxO81oupK6a1BJ/aqWMLAtja9banq0AotBlLFbStYHMZ5tkeYlB5OMmmcxiHLed7ZJzTXJS5tBXDBISp5bG7LQcDnT8jnT9FgbJFimFtsThsG5tsfruzGlUri2SV5JOoG2QPnq2oDtUcLBNMO1TfzVNq/+v/53+lHBXJJjCIOk1CDllu+QFQnLDZ/zHZ+DqOB8J+Bc26fh26eNLk+Kl9ZcmFHKA9vSzLaGj2Mb/ODjSzyyVGd9Jo1xkhl1qMK+3o95dl0rIB+K/vXj7Cix6UfFKWuIk759D2rFHK6hKN3Ksg1BIfVn8yDLnp1xQpJVuJbQ7bkTY+9XiTnZ/rq2ZyBEOvPmulst+eTrHCZxD6o2/X6Md7o7+rvJzrcwDoW0BOKolHq/HvKdk64XaRPP959v32Xkdi+58IWag1T6hGabxtGX7XBSnuv4rPWiGcgPVts+47QgKgoCyyCVkg9fnOMjF+f4wo0eV/ciLs+H1L1jXYyNmQHoc2t9tkYaAH2m5fPM+RYgeGlzhGlqQPN7ztQ51/b46nqf28MYE7hqGdiGwVPn2vSnKWv9mLKq8C14aqVBUhTEue57j9OC8x0fz7MRQpfd6wrOtT0GSYltm7iWXvT/yMML1Hwbe7ZQfbWQ2IZm0mz0U6ZZyeu7U/71l25jmlrhdqkhWKi7TNKKrfGUl24PKWWFaRj0o4I4L5EKPMvAmWnEzPkWDy822Bvn7E0KhNDJjGdp0G675tFwdZuvrDTjJ7SgHdp4jk071D87o5RhpCsfSamTnXJW2Wk6gsLUbCvDEMwH2vDUMjPismKl4eJagt1xclSNyapj89FDdZQjjIwJ8zWN0YiykkpKAsvUjsylnCVV0Agt5gKbq/vJkZbQYUj0/TU9gzqKvfg4I3KBy90647RikJQoWZLmFeNcv36Crm4JoOZqEOuFTsDruxGeURLl+v4VMErg1TwmOUGXN4GaY+BZJgdRziAqqCrFIJ7iuxa+Y7I/yfmlr2+zNoyppH7PNccg9CxWmh4XOiHPz+jPo6RCKKkTT9dmLnRYbnhc258yiAvi1OIjlzpYhskkKZAYmnGmoO5bnGn7dDyTtEgZxRnKtrQVROBgGtpFfBCXGpsySkkKSVFp7zfDMXFMGCYFaT5GmCb7k4zXtscMk4K6Y/PYSk1r7kw1ILc/zbg4F2KiaPoWz2+MyIuKnUFCVmmF86Zv8+GLc1zphnz+Rp/1fswrWxPSsmS1HeJaJvM1n0rpQ1TNs/Btk7bvMEwKap6FVJI4kzRck6+tD3lpc8Rc6OK75qlN+eRBSymlDTqnKa3A5fsf9VhuBjwxU3h/eMm6S9hOCO2htzfJTlUAnjjTODroaZ8tCGzzKCm40g2PfLjuhwm5U0fsUAesHTinVJRPxslqxDjJ6cySuDuve6/XPJkEHa6x9zqgnnydk0mcEOIdV/34RuL3smX2u4l3k51vcRiGcddEvldP83Ay9KY5twfxDIQo6Da8I7rmnXGkm5MU3NqPGCQ5nmWy2HC0zxanTeL2JhmuJZikJV9bH2glX8Og23C5slCnOaPCXpyvIQyD853giL64PpOQLypJUUrWB4kWUTNNVpo+64ME2xQ0HJNhkvG120Nu7DtkuV7o47Rga5ho+mzLIysVW4OYSVYQ2jat0OKJ+Qa3DiKiXOupTDNdBat7Fg3HYr7uMkoKxnHBKNbvdbUTMC0kN4cT9iY53/vwHIYhMCyT7X5Mp2bRj3OipODWwYQrCzUuz/nUA5ezDYfNUcZrOxN24oqirEiLDEMIPMukFVos1Fx6UcHvvLGPYIE///QKZaV4dWvA3ihjkoFtGnRbPk3fxkDg2CWLTZe8kAhhcKkbstj0EEBSSP7Iw3MM4pL1Xnpk9WAaMz0aISiVBkHnOuNhvu7y0csddsYFZ5s2X1wbsdAoibISx9C9syjTlZuTIdEVonMtj7OdgK+tDxmmmsVkGIqaIyilBrouNzwe7QaM0kqrYFenWzeVhLyStDxBAEfifbYFSVGx1PBYqMOLt/uMkmOhQwUks2Tmje2EbF6x2g441wmI0pxaUXEQF4xzPS6vjhM1z4SzLZdL3RqTpGQqDCp5CEIGu1LISjFJc67txWSlriS9d6XGoysNPNtipRXy5HINx97m89d77E9znVS6Bt2my3I74APn6lzdmzBNJWlZcKuX8OhSg6WmR8PT/k2GVISuiaUUuYTQMZHKRliCpBQ8vtTEsARPLdeZq7m6fb0/YRAXWKZJ4Nh0QotSSqKkJKmUtixRcG1/ysE41dpQNjR9m3FcEXgWL26OeW59SCd0qCrJaztjPNtimpec79RoBQ7fcXGOp1dbGvze8tlo+uyME7LCoOabdEKHx5ZrXO6GRwDabs3lXNtDKclnr/cYpyWfvb5PkpXcHqWEtsnZTsD5udqpTfnkQWuSFOyMEm71tN3Kd1ycO0p04P7tk3tVAO5MHA4rJoftHgClJFf3ovsCgO/0CLzTRuJecUrvJ9X4rpMU+beDQ7lT5PXkuGPTZXWUxL0Tqx/faLyTxQzfTXZ+D+KkAumd4OLD8vCdk+GwsnO2fX9zuKOJWlT0lPZp2RmneLbJWj9GII68ti7N+0ySktt9TVuVSrerFhsO33Fxng9dbGvn4RmzqpKwH+VcmJUM9iYZdd/GswwWag7jtGCQFDy+WMexTYZJgW0abE1y8kKyPzkgdC0qpQhtg7bnsjPN2dnq8/1/9bsA+M3/5dcolcmwzLg9THhje8J83eXSfMgz51v0ooJXdyZ81+V5Hl+pE2UVb+xNeX5zTFyULIQO3bo+TfUnGZuDBCklDcdGKUVVuihlczBKmeSSflxwvRdzoaMNDedmBoR704xpnDIole7XoLAtQcN3QRhsjaZUCr5084DFhsvlhZCNQcxaX2t+ICULgYUwTQZxwdLMjflgWjDJKopKsj1K2RymJHlFWTV5crnOIM6JEs3Ycm2DSil8U2Arg6SSzAUmKy2X77gwh2sZjLKIYVbNElqXPDfoRyVJWSHEMYjY4BiDI9Cst4YtKKQiziocS1uOmIai6dnUPJsor3hxJ8IQkrmaQ5YXDFNthHpYLUozyCyNg3ENEMLAFIKDaUWnJliqO3ylOsYTZSfuA3R1qJr9CCVJK33foW0yzasjlWbXnAkyuibnOwHLLR8hEvYmCZYpEAIcQ9H2DeKiYn+kGVam0EagrmOSVQYfuNDiuy7P8eW1IZVUGnCLJKlg2Tf50e++QBi47IwifNtAKYktTPrTjINJxvX9CGHASs0jzwu2pyVRWtLwLZabHlFeMU1LgrpJp2bT8B29uTomLc/Gt21dcZU2Ky0P3zbYGmZsZRlZIbm04LHS9uhHKdNcIgrFV24NeHq1RVopLGmQFAUowY39KVUlKSqJoOTSnDbsXWoGPL5SnyUFWjDvXMfn6dUW28OYOJfEgxHiiVU8AQ/v7rLaabPWi/jCjT55WVFzTA6ijIZrsDPNmCa6PXhpLqQZ2nfpeB169X311oDNYUyUVyw3fQZxTl7qFs2brYeXFwJagUUnsO+qmpzcMD3juGJz6CR+EgB8SIs/vYYmtH1t/nkvQsi9KOOHrbXDROpeQOQH4VDebNydba9vBWbn3XhwvJvs/B7GnT3vO2mKpyiQRUVWShq+fd8JcciQyIqK3XEKmFyYC2iH9im3YqUSJmnBWj/CNg06gUVSKJK8wrUs8kpXUkzzzXQg9L+XF3TFZ2uYcGFen4CWGj7ZYkWUlVzdm3Iw1YKAeVVxYa7GmYZNicCxwE61UmrDs8iVwTAuGcU5hgHzdYf3n+/wsUcWeH5zRM2zaQc2H7zQ4fp+xHPrQ9KyoigUDd/hA6stkhL2pym9acHOJOP9qy3++Jkuv/riDi9sjmjXXMzUoB8XpIXi9jBjcZhwa5Boz65ehFL6M2gELq4J5zoBvm3qapmtzQdNU/DC5gjXNjEMgWUalEpimoJMKuq2Vtb1LEUj9Gj4NrcOEvYm+UwXJccQgrVewnes1vFNk8JSBJ7BuXaABfTjnHFV4NgWy02fD12c50c+sMKvvrjHwTTl1v6UxbrLUsNlayiZFDnVDBPTdHXCU8zo4VmlJQZe25mw3o8oKqX1caqKYa5TkMDR6tFxoahQNDyTS3M+vh3y8vaUUVIcscAyYC+Cji94/9kG40yyPsoQJsR5ydZQJyyHSdfht/Zwu2n5BhfbHuNUtwGjtNAYm1kLL7A0/ifKlL7/smKYFMiDmOsHMVIqbCSWZeDbJmmhGEdavChwTCxTYM+YWOYoZmfo8/zGmKs7Y271Y4ZpjmWa+Jaiqgxe2p1QVBNe3R6zMyowDEjLiut7E4ZRjkTgWzp1dG0LZIH2ptLvtygqhClwLV2FrGTGG3sRjy43CD2TZ863iNKCUZKxOUwIXYvQM6l5DralK66twOUDFzrsTwuNNUFbX9QsE8cU1ByHdmhTScntXsQwyjnXCVid8xGGtkO5tn+62hG4Nk+eaVJUii/eOOAgTfmeJD5cNBDA19aHrPVjsqJivuZgIPAdm/PtkG7NY67m8n2PzGNZ1t06XkXFzYOIjVkJ72zbxzYM9icZa72Yh5fqR0nRvdo/Uko++eo+1/YnBJbFxW7AcjO4b9XkyEn8DgCwY4q7qi6aAl7RjzOu7Rl3XfN+lRrfOXQnP10Jeqs4lLcy7s4k7g9avJNp5/AOT3Z++qd/ml/6pV/itddew/d9PvrRj/I//A//A4888sjRmDRN+Tt/5+/wb/7NvyHLMn7oh36If/pP/ymLi4vfxju/d5zM7lGKr9zD+POwdHp7kB71nu8FeDuMvNIMhYbv4Fpazj0uKt2vVrA3zWj7Dr0oxTYNikrSDj2WLH2tQkrSQls+KGWcusc7dSAOe9JFpRilJZUS7E9zLswrHlqscbbt0QocTAMOopIky6ikzf40peVbPLHUpJgcU887volpuZgigxmL6VK3weMrDRqBQ7fmcm1vSpJX7IxTHl2q8d0PLTDNSoSAD11osznKWetPCRwLahrIqpTAtUyivJqJECps0yTJKyqlMIUEKTGBQZQzjAuKsqJTc3ikGzJfc8ilIsoly02fhmviuS6W0NUwz6rohPoU348yXNNgY5DhWAWmAUtNjys1h3boYRkmr++MifMKJRWGCZOsYJRX1GsOyjTxLe3BNYwLDMMgqbRfxfYoJ8tznrs1IisK9scZ+xMNYG0HFmmaUxRqZpaqMUpZKcmKCiUESabp7GWlmGRam8UxDeK0OnJWj0pF3aqQAgxhIDAIHYuoqDTl2zLICnns6wVIZbDQDFg0TUJPa8X4rsX+OGGhZjOxK0bxMRXeMMFW0PJN1ocJVSlJiuKIeq4A14LFlkvTd1jrR5BL0hzWexFXljRDq1ICYZiYpkBKGKYlJbrd9chijZqn2VMGAiXgK2tDvrw2wLNNelFOzbHwjIphWrIf5/zOGz0udnz2JhlxUek2oNDilaOsJLAtWqE7syooceyMVujT8nTlNC5KXCmYZjbjdKqxVYbJ+XlNQ84KRZyX9CbaJLftu6yea9AOXNb7CWdmIODL8yGjtOT17TFLzYCPXuowTir2o4ILZ1w+9vAcX7je57WdCYZpYBqC6wcx5+c0Pb6UirLSSdjhOpJXimlWMIpLpoPoaM4ppY4qbZYhsByTUirOdQLOzwV0aw6/8eo+O5OUz10f8N0Pzx/peB1Wo9d7Mf1phm0Iuk2PJ1Ya7E9z0kKyN804X4b3pXKDdjW/djBlb5wRFTGBp9e2xYZ7z8PdnU7ih+vhyTZ+XkrduheCYVLMHNFPV1gexLo6XJ8fBEQ+XA/vtam/0/Eqv9t4qwnMO512Du/wZOd3fud3+LEf+zE++MEPUpYl/+Af/AN+8Ad/kFdeeYUwDAH423/7b/Orv/qr/Nt/+29pNpv8+I//OH/uz/05Pve5z32b7/7ecbLnfb+TwEkPq9u9iJu9iIvz4V1AZTheCKSUTNKSaVYwX/N5qKu/bOfL8ARbQBA6FpM0Z3+SU0jJQsOjN834wo0+59rBERDwXl/uw550t+bSrWuTz5P3vjFIibKCuufimCam4ZMWBVujXOMwBCzMQIMAt4cpiaM/kw+eb/ORy/O871wL39VtqKv7Ez5/7QBTQG+a8d7VNo8t1VntBAzinHbg8sUb+1zdjTAMxSPdOuc6Lq6tPX2kAqnkzCRT0g4dQBC6FpuDDGvG1rEMge1a1ByThxYCEqkQhWQQFRhCsdSq8f7VJs9vjDiII0aZ4n1n6jyxXGdrmHJ1b8LetGCalviOwSQpcS2DP/PeJa4u1fiZT8WMk2wmmKYB3PuTnDMNl3Mtk05gY5kC1xDc2E2JU4VpVWSm4tdf3uXffX2XpaZFnBVIFFFakRfaP6zmCILA5GwjoFKCQZYhUCw3fDxDsTnNSPOSvIKsVFSGTrhOcsgFGtBbSkFWVtweJOSVFnsMHRPPVBxESjt0GzAXmCy3XK7tp7iWYLXtE5cVkWvTDFxcWyJUSlJUpJVuLwkL0gLyXDt+F4WuJIrZNbt1m25DC0l2fIdRkoKCaakYxjl136aoKkZlhW0I7YotBIapaPo2dd/i6dU2gW1wq5+w1tPYuMCxqKRiYdbuvTgf8LmrexxMS67tTlgILQwUwtDAUlMohKHZVUWlMJWmTveSguWmBsqWUrI5zpFU2LaJlArLFFgG1DyLNK8I2hZfGwyYpBU7U51MKwxMy+S/+egFbvYzorxgqeGz2HCxLEEncKmU5Fy7RrhksjXOONP0OIhKbvVjbMOg4dn4jsVqO6Tlm2SFZh2BTgROqQmHLg3fpJgcz+OruxOuXAxmZpgeK01Ntx8mWoRvoWbj2D2MQnC9F/HhskO35pKXkm7NRaAPUA3fodvweOZ8m7pn8er2lH583AK6VyJyqABf9yyuzGttq8Cy6NYd0lzy/MbwngylB9G/u3WXjX7MJClY68VahuMe6+qdbKi6d296+IPW7JPXOSkmeHK9fKfiVX438XYSmHc67Rze4cnOJz7xiVP//5f/8l/S7XZ57rnn+J7v+R5GoxH//J//c37hF36B7//+7wfg53/+53nsscf44he/yIc//OFvx22/pXjQSeAwgcmKimt7GZujlEGUc74TnDKfO8y6Ly8ETNOSl7bGOKbuU+dVDc82jhKrcx2f1Y6PZcAnXt5lb5JqYLBnsTPOiLNSgxpnrYP7AajvBf4Djk5Mw6SikJLFps9C3aXu2ry2M2JrlPLyzoQgT/kvZvdvmYL6DDPysUeWePp866jHPo5zvnxzwK1eTFVJTMvkwnyIEIKPXOpgCEFZVXx1rUcv1rTwaV4SDyShYwEFZ1su4yRna5wyTTVtd7VTY3uUkEj9Ps41PXzLQgpF4NjkpeSFrSlC6M+3EQTUPZMri3W2Rym3DiJKqXhpc4JhCLoNh4br0I9LHEtgGhqw+/puNDNE1B4/aWWAklimgWsbTDPFmbbLlW6NP/XeZfYnOZ+9us/WIMaeVYhqtkE/qUilxBopLFNhG5JCaCuKQgrmag7LDZ9LCyE3exFKVpjCwLcN9ic5sjLo1nyKsmRSKKSUtAMbb5QwzsC24UzTYz5w2JjmTOOCQZJjCkHTt5gPHZRUOHbCKCpwLYFrCw4mBWVVsTFIWR8kXJivcXk+oOFZbA0TKqWw4gyV6aTBszRYFmHQn2qdIqOCpabDpbmAp1fbxEXFRi9hqQlxIRkmOaYwEMBSw2V3kqLQ76HpGUgJlRQEjsEgKkBWXJyv86WbI0ZxgWuZXJwLMM1ZtSorubLg8ek3hKbFA0oJOjUPYQgCS2AYJuMkY5RWTHPJ1jijE7qcn6+R5iVzgcskK+iELr5t0fBNfNvCMk3OtTXF2rUtbKEroKYJyw2PKC/pBC7dhodhmjy+XKOoFPVZJWOlFbI9ytgaJry8NeJDlzp89PIcKMUvf32Lg2lO4Jk83W1xpuNTd01cywK0VpRvm6x2/FOn8MeX61zfi9jYODhaM24dRJw9U7LYcE8lIIfPU0rx6GKDawdTrszXqHkWzBicCC2fcFhlOaRoX9uLjmjjV7rhEQ7xzkTksL0lhOC7H5rjQxd1ojTNKp7fGN6zGnNyvbzXY6udgI1BQpJXuqo0H95zXT1cvyZpRd0zee/Z1l1VpLdSxbhTTDAv5ZHFxTejmvFOagW9nQTmnU47h3d4snNnjEYjADqdDgDPPfccRVHwAz/wA0djHn30UVZXV/nCF77wjk524MEiUFe6IU3fZHeUzCTpT4u+ncy6m56l21Qz+fh24BwtXFlRsdaL2ZtqBkZWVlzfm+JYBudnRoG9acE0K7kwFzKI7zbYA61m2vJ1xWWh5pAVFY0ZUPDkiakdWAR2HdcWdEKXy/MBlxYCvnSjRz/KSaPjksJTZ1osLneYq3k8uhSS5iVFpWZUWEXTswhsA8s3OdfUNNm8Uqz1NP5kd5bELNZdfMdEKY252BknLEiXGz2NlxEKPEu7Ra+0XDqBQ1aV3NyPeWMv0q29wME1K17bjzEE1D2TSkKSl0RZxfX9KWUlqYDdSaa9kqRia2jScG2W6q4WWfQtBnHO5ijhN17dpeObJHnJJMnIS63Fc7bl8fhSjYM458XNMb5jIaW29KjQ+AHPFlxZqHF1P6ZKKwolSTJF4FgEtqQoFbZtsBDadJs+dd/ikeUGdc8mLipkVTEpSiZJRZwXmCbYpsW5dsDjKw26ocVru1Ne3YkQAmxb2xagFA3PpumZtHyH5ZbHjV6MMc3IFUSJIpMFn3p9n7xSxLnePHqTlA9f6GCaBtNMstL02B6l7E9TlFK8Z6XBYitgITD5R5+4AWgcUMM1efJMkyvdOs+tD/E9iyuLITXPZnsYsz5IiCtdvUlLSX+q8U+WAaZh4diCvUnB1jAnl4rtccb6MKIoSh7q1vnr37VKoUw2hylVpSiqkuWmR1oqQttkvumwNcgwDYNm6BDnktBzMEWFVNoypBSwENpYTY9q5v32sYcXsA2DUZqTlhWLdY+VdsAwLpikOf1pxoU5n6KUxFlJheKRBV11+D9f3Na+Tw2P951rstz0Odf2uHng8NLmmLVBfGT66c4OK6Fr0fYtVto+07Rid5STS8mFTkjDt1hseKz1YzaHKWfbPg8v1ikktEOLiwu1ozn36taAjUyQVpIrC3V+4PGuViOevU5eKb7/0XneFzfpBDbTTOMB00IesZROAno/f6PP3jil5lkUleKhGUD5nonIPdpbzcCh4R9qgd3fn+p+CYBnm5xrB0fXu1+F5c7NuO5Zp675ZhWbe12n5dv04+ye6+XvJr4ZraBvZrL0dhKY3w9tvN83yY6Ukr/1t/4W3/md38mTTz4JwM7ODo7j0Gq1To1dXFxkZ2fnvtfKsowsO3YCHo/H9x377QilFNf2IrZHMc1ZuXi1c5p+fmerq5JQc02eOtPiiZUGwGkGg2eTFRUb/YjNYUzo2nzoQoeoKKkH+mR4pRtye2AeLTqHmkC2Aa9sT9gZJXTrLrf2I37r9X2uzNf47ofm7jox1VyTV7YmvLQx4sWNIY8v1ZmrOTx5pkFRO54wF+ZDgprP67sjfuu1XUZJgWOZrLYDLi4EPLHcoJCKKC1o+g6jWLNYvnjtgL1phmsKxlnFfN1jIXRQUvLVjTE3D6Z0QpuaZ1FJjV/xXcFy0+XCfJ12YNFtePzKV2/zues546wilxo83G24WKZB3bPYGaUYCNb78UwNO6OsJK5tYAuNvRnEObsT7Rl2tumSFJKqquhPJHlRslR3sYSYOZbrDb4bWliG4usbQ6QUJLkkdA1NcVYV9cCi4RjsTDXTyrehKCuyCipl8NB8iGmabA1jtiYZG+Ncb5oNvRkXlURWYKEQQhFlktAzCV2TRuAgDINBqtifFkwLvaBvTXLmQ5+lhstyM2BnnHB7lDNMK1xTt/5cq0QIrQHUi3KyGXDZtkzaoUPdM7jZr1BKMhdqgTkpFZUSOI7NQ90GkyQh9ASjVGELcG2LKws1fMfkvWcbDOICU8BBlCPGBm3fYT60eXSxTrfhk5WKjWGCMkAZUHct+lGBVIqDac72KKXjGVyflryxN+UXn93mL37oLB+52MZ3bUZRxstbI8qqh2mZuIag7ppM0pz1XkKuNP2sFdg0FVoTWeoqmpCSUVKyOcx4aCHgBx5b5MbelE9fPwBlcH4upFNzycuKG/sRUip8x+LR5ToNz+aps02evz3k6v6Ug4lW3N0cJDi2wcMLNeZq2oA0zSu+fnuAQvGRSx3SvOJgkhG6JqiZ+GNRzcQz9ZxzTMEvf32LzZnFxfmOxrUsNX3yTnA057aGGWk6pOnZlFLxoYttWqGLlJKXt8b0I401cm1x6t+TbZ+TWjnJTO9gEOVM0/JUBedeicj9qgX32yjfLAE4yaZ60PZ6twjgaQmQtKi4PYhJ8upIP+heFZt7Xed+SdrbjW+0FfTNxs283QTmnd7G+32T7PzYj/0YL730Ep/97Ge/4Wv99E//NP/wH/7Db8JdfWvi0Gbijd3pkX7FodHcnWqmh/oQNc+g7tk8tFg7opDvTtKjxch3TBZChxc3hkRZxULN5ZGlGluj/Giyeo5110TeGSeM44Ln1gekRcWVbo0oKzU4tJR86EKLbl339Bcb3pHq6t40Y2OUcDBO+dy1A9qhzQfPt1k502T36e8AIPQd+lHK87dH3DiIyIqKhYbLOCmo+ZqF9fS5Flf3I272phrwqmKSoqI3KbBswXdemuPSfIhhGmwNtFKzb5kMkgrfs2l6Do5lkBcSqQShIxgmJUk61vRmW5CWmso9X9MnaSEl++OMYZTTj0ukUuyNUrKqwjQNap7Ndz/aZj/K+cTLe6RlSZLnGEKR5JK0qMhLiSFseobAsQzagY1llFiWxSAu+fLaNgeRxDTAMQTzdYckLRkmJaZhIi2TcZ4zSEqyQuNeDAMcqdif5qx2AlzbYrcXI4FrO2Mc02SSVUyzirqr2VzCjNkb6laSUpLQEawPYrYHCUleYiGplCLPJVPL5Mpik4cXArZHKb1JSpybnO+EXF4ICR2LpKgoSsnNXkwOOEDNFkzTnP/wwg4HkwzTEIyTnGlWUkrFSsunKBUvbAx5dXtEy5u1X1yDwDLYG6dMs4r9SIPpS6VYqbvc2I+o+Q5zDZ9Hl5vYpuLW/pStYcQwqjQeRRrUXYPcMljt+HQCh2EiKCpFVkh+9cVtnrs95EPnO/ytH7jM9YOYju/QrnuEtsnupGBnnDNKSnzbxDdN2oHDmaY7w3ZNqZTJ67taZM82BS3fRAnB7WHC17fGHEwLkkLxg08s0fBtNoYJiw1vpmlVaUuVpk8nsNkcpGwNUmxLsFBz2BiniBy+FA/44GqTxYZHP8rpxxlf3xhSSa2V1fRtTMPAdSw+uFQnziTTGebHtQzkjBpelJK8rKikJCu02rk0DW4/9UHySlIZAt82GGclrUoLeNZck+c3xnzpZg8xs7xYqHvsTzOWGx5zNeeuts9J0PBqO2Bvkp4CKB9ueufaHt26c/TcU4zUOyjtJ2nk926f3z8BOMIT1l1WO8eaQIdxsuJxZytqse6wPU7ZG2sG62rHf2DF5l6aQL/XlZR7xbcCN/NOT2DeTvy+SHZ+/Md/nI9//ON8+tOf5uzZs0ePLy0tkec5w+HwVHVnd3eXpaWl+17v7//9v89P/uRPHv1/PB5z7ty5b8m9/27ilM2EaTJONH4lr6pTFM2HFmuca3t8+o0DrvciQsfGOQFQPExCriyEnJ8LUVLyydf0qcwyTcyZfPudk/WkJlA/KljrTZmmBXGhFXfPdwIObg+ppNaPEcZskqvj+z/b9tnsx9zaj1jvx/QjE6lgvu4R/k//GtMQrFYGWakIbIO2b1M4Jistn8tzNbp1h6WGT15JbhxE1ByHaZ6zWA94bWdKqRSy1CdVxzLxLJOFmsOji/WZK7MNSrIjUpJxBUJXqV7enPDy9ojeJEehCB2LlYbB+U7Ixx6d5/mNES9sjrm+H+NbgBCEs82h6RnkUmEqRSv0uLxQ49WtKS9vjzGEoj8pUYYgK8E1TVzbopSCXFZ4rsVyw8OZCSxOspmxo9J6N4Ok0EmSEnhoiwfXMhDq2AW9kpBUiq1RNhMhFEfu4JMCVppaANExtBGpKQRVpZOBUklsy8CzbfamMcOkpJSS1bbHNJO4pgKl2O6nvLEzZbOfUEhJDQvLgE7oEboO86HDINbspZ1hhgByKdmfFORVRqUEDd+izBW2ZZKlJb1pTn9aEDiaou/aBiuuSct3sW2Dl3cmdAKb24MY0zTxLQvblNQcc6a+LPg/X9qmqCp6cUHgmAhh4FmCflyx2vI42w55z9kWG4OIFzcHxFl5ZBuwM1R8USn+jy97vLw7oeGYdGsOSV6RllrXxzQM5kKblXaNudCklPr6SSnYHE5xLYPLcwHdZoBtmjimFmLsBDbDmqZsP75cwzBNzrU9ru1HvLQ5wjK0/EM/ynhhUxIXuo3WCR3+5HuWeH5jwut7E6pKMsl1a/RcJyQpKixhEtiCM60atp1weT7ku67MsTFMiQoN0FdoVudCTZvyvrZTcTDN+K3X9jVua5pT8zw+87/9Iq5t8GRSUvdMbvUTPMvks9cOuLEfsTGMGScFTd/mwlyNmmdqGYWZgvGd+JajU3/HRyr9Nz0UK0RpXNXV3SlfXR8AnFKCP3zeSUr7lW54T8HVt5IAnARDb/RjNgYJ59rH6+S9WlQnzZO/sjZgf2bfETgWD3Vr3B6kb6li881MBr7RVtDvB9zMtzPe0cmOUoqf+Imf4N//+3/Ppz71KS5evHjq9x/4wAewbZtPfvKT/MiP/AgAr7/+Ouvr63zkIx+573Vd18V13W/pvX8jIcSxzcTBNOHqbsTrexPans2HLnU41/HxHUuXkw0D39Uqxp5jnDbJU8xOe2p2ooHFhsbtLDZcrblxn8l6UuTQs+pIqRjEJYsNj+WmNzOzNNkcpQgh7jrVPbxYp1tzqGRFpSRFBS3fJnRMNvoJEoWU8NBiyHvPtWmFERc6AU+vtlhp+ZRK64GYQnF9L0BKyRlba/zMhS7XDqZM0pJK6UXVc00e7tb5E+9ZZrnpYyD5N89uYgiDxaaHaxoMk5Ivr/UZxjmjOMe1LXpxgWXA+lBv4Oc7AUWp20p1x2Kl6aOEQihISsnuJMM0TfYmKbYVcmEuYJTmFEWBbetWYVRUNFyb73mow+s7Y17dzai5Jo9ebLLS9Hlpe0wvKpjkJQ1X0AhcelFOaAlkpatwCg2YPdt2SXOJMAziLCeZmXnuTXLdWjIEgS1o+xbC0C7yRVlye2bFMUwq8krimhp31Y9yqkqRFhLbFFTSYK5mM4pzTMOkHxf0kwLTMulYFpcXajy0WCcrK2zTJKskgWvzXVfmeXFjyDAuMU1BnBfYhoEpFOeaLt2GyyDRxPJxUrA9TAlsm8e6ATd6BgZQ920C22JvmrE70qateSFpBBZVJbV+jm0xSrRat+/oBN5vBRgCbu3HlJVikJRcWTLZnSTcOIjYGuXkJXR9aPkuSWXoJC1JKSvJMJE8ulhjnBa8sjVknFe6GqIgLQoEDq4NjmVqDNusWpZV8NHLc5yfC7lxENOPMh5bbvDocoPzHZ8bBwm9SFepXNOglBDYBm/sT7kwp1tJF+Y0fmal5TNf9/iBx33ee67B19aHXN+fMk5LQlfQ8hxKJSml4Iee6BIXipprMkn13zYtdLUOIMm1vhVovaG9cc7WcI8rXW346dsmtiGQSif3vmPi2bam+CvYHMaYwqAZOHz44hzvO9ekkLwl8bv1EwzNZ1ZbrA8SvrI2oOXb7I7Tmd6XOqUEL4Q4EvMcJyVSxviOYHsUM820q/qbGXHea63KS8kk0Z59u5Pja9yr4nFonvz8xpBxevjZaUzbyQr3N1Kx+d0Yf34jydPvB9zMtzPe0cnOj/3Yj/ELv/AL/Mqv/Ar1ev0Ih9NsNvF9n2azyV/7a3+Nn/zJn6TT6dBoNPiJn/gJPvKRj7zjwcl3xiGYWGuO6KTl8ZU6B1OHl7bGlJVkbRizPNDAwMNTkmvpkxdoszzHPF3+vXkQ88beFAQzp+UOi02fsy3/CJR4P9+Xw4kjK90KiwutvnsQFdRdi0KqGQVdMExOm/QJIWiFLh+90uVMOyC0LULXYHuUMo5zelHB9iTjA+c7CKENKW/0YtaHKY9065yfD9ibZEySko1BBMLgSjfk/efn+N5HTMZJwVov4jde2eP13QlLjkm3ZrPc9Kl7Fv/ppR2eWxtiCPjQ+RaPLNX59Vf3NM3c4Eh/p+EabI9TlBK8sDWiHdp8+OI8vTil5jk8vlRna5SwP8mJi5LQtTENWGz42j1bKjqhg+/U+OD5BjXP4XYvYpxVjLOKKFd4lknoWpzt1PhzT6/w/v2IT8/tEWX6PUgJTT+g5dmMsxLLMOhFOUKYLHguFzoeEjiYZGxPUkZRqTc7qcUPV1oaa3N9L2JjmLA7TjEExLlEKokloO4YBJZJKSsCWxA4gsA1CVyDx5ZCojzAMQzWBhEIiLMS1xI4tiApS+quzTQt2I9yokxyZcHj8eUGe1GhDSYDi4NJzLRQrA1SzrUDPnShwSQt+PyNHlJpXRzRT9kdJfiOpVuessIwBI5p4FuSrFLEWYkwBA4GpVQUUjHKCyoszloGT5xpcbZl88vP7zKMUjzbojfJGEYaaG+gWVCWafLkSoteWvHMuSaLbR8hDC4v1Fnt+Hz22gGB41CUGU3f1mJ+rkMvzrWxp5rhdYoKyzRI8oJCaguYQZwziLXe0XvPtljvx3z22j77k5RRnHN+rkbDt7ixn7I1zhHA48sNPnop4FdfVGwOIz756j7n5wM2hymDuEAqLSppGy5KSDzbIi7Ko0Tna+tDXt+ZsD/V7MMr3TYo+NrtIUrp7+FyUzOyLGGSlyVXFlpcWQj5Ty/vsjXKWGm6/PClZeqexStbE3bGCVKC5xjMhd6RYac5228fJH532G7vRwVKKZaa3pHn1aG0xpmWxkbeqQR/LJkRc3Mv5qXNEYFtcbbj062/uRHnybizWnSoGn+Shn8ve4qGbx8Box/q1u5qf30zwMa/18aff5DaTt/seEcnOz/3cz8HwMc+9rFTj//8z/88f/Wv/lUA/sk/+ScYhsGP/MiPnBIVfKfHySQD0PLrawOUUqcM+C4vBDzcbWjgaamZUCe9W45UlGeqoVd3BefnAhzLoBO4FHKMbQgt5V6pmcS8d2Tw+WbgP9cyeOMgYhBrWbkzyz6CGQ7Fd7AM2Bwdl4fvlTBlpeSlzSG745SWzPix//L7QMEv/9JneHQp5HPXB2yOUtb7Ce3Q5vXdMXklKZX2z+rH2nbh0nyNuqcrWr5r88hSg71Rylo/4ub+hMAy8RyLuutwbW+C5wiiVAOUQ9fm0kIN29Q4Ct8SrI9SfEvgOTbDaQYKRmnFxXmHP/3+ZRquzQtbY8qBdpBu+g4X5l2W6h7TJOfafk5RSZZbPo8tNfjT713BtU1e2RrzO1f32RoW2IamcD/crWmQrGPx2HIdpeCTr+2wMy4oK8Xl+ZBm6GBaBnkFF32baVYipeJmPyEvNVj0vWdaTOKMF7YmpBWc6Xj8mfcu8+z68MjkNfQs6rZJWpRMU+PIu0oZQuPASkkzdHBMS4sAJhXb44SFmsuTyw2+ut4nUpK8MhhGJXGecKYlNTOpqNjox7y6pU1f37PS5D0rDdYHMZsjk6QoSPKCqwcxB2lJ07WwTRNhwH5UsN6PUVLhORWXFgIWGx5nmgJTwO1BSj3VbaooLcmlPFJfbjgWTd9kkpW8uDXCtlqcbXqMk4JhrFl6cVFpY1PTpOma1D2T26OUQsKLW2OiXIN6Ly8EPLxYJ8pL9sYJc5GD75i8f7VD4Jpc35syjLVNQ7fhUqqZEGQi2R+nPNytkWQV670I36rjmBqsq6RiYxATpSWbo5QnlusUFTT9Q10tl7hQbI0T+nHJ67sTlECLAkpdpemEDq4taHoO46wktG0CW/CJl3b55Ot7pFnFhfmATs3lwlyIUopbfe1N5TsmP/REl51RytfXhxxEOVf3pnh5yp/74x9AKsXHf/lzuJZBUWlLFMMwONe5o+ryFsMxBWku2RolBLZJzTVPJRVXuiFXFmv3NMc8XBtagcVLW2N6cUFPFsw33NO007cYQmin+YeX6kesrzcT/PtWVkJOeh1uj1NWmtoa6J2oP/OHJd7RyY5Sb/6t9zyPn/3Zn+Vnf/Znv+HXiyKOTjQnwzTB806Pu18YBvj+g8dqttWUgyhldcFjteOzMUi4uZNSVopNu+BsJyBNBHNewEfOd3lysc24TDVlte5RZoJeVM3aU4rtXsE4qbi+NeRaI+GhlYDHlmsoFDv9nLrlMZlINoYJ+5OMbt3jTMvjjc0YLF25WZ0LUKWJPOEAmRaStb0MS9nYM3NQz9FuxlFU8UvPbbM/SVlu5SzXavgnfHHCUCs8708zrm0nrPUSLnoSNdKb8uOLcySpiSoNzrcDhFIYhsDDZn+gS8ttJySNBXVfYEqTwUhykCbsTbXT+2Basb6rrSneUDHdWghVwmgscZXD8oJJ07fYGSecb4a0LA+pYH+SkeUGTc+m61Vkc5KCgrmaw8444+WNiOV6iCdckkRQtzyyVGLaJq/cnnBzOMEwDK4shjy11Ga1XefmXs7GQcR+lDEclYzGiobjc6Fj0q7bvLYz4Ss3h2z1c+qOwfXtnDSBrFDIlkUU65bW2Y7Hxx6Z58W1Ka9sTfnstR4mgqoQqKZF07No2CWm0Bvyl24NeH0rosgEV7otLnZ8GoHN1Z0JV/ciskoR2or1ccIgLjExcAXMtVzGE8UrOwOkVGy7BZMFRW9a0k8qXLNiLvCwlMHOQUVUlEySgt64YpiUmpqcghAQZ5KabZDEFqVU7Oxn9AYFjmlS9y3qhksiFJHMKSsQRcUbt1OijsH7V1s8c6HDCxsj1voxnmVQ1SpMW7E/1knmNAbXMskKiShNRmOJLEyqTDBNFaUqMAxBx3UoM5OFuotQksGwJCsrDvo5z14bEwYmlqmopGCSlJiVzSPzLRbqPj/48AKv7EQMXcnmQUnNM1jqeHRCj16U03Q8apZHFMFWr2A4kWyaOS84EZsHmpF3aa7GS1sjxiPJvlvRqdmkmcIXgtc3Ex5bqnOuVgOmXJ4LOT8XsL6X8+RiSF5IDdbOK4o0xg88lus+g6TkVj9CSkmaQd3yWAxCiszg2t6U7QN9EJGBwetMmSQl/aRgp1+QpwbNQnFpZh9zpdPktU1tkzIuNEtzf5qzGNaQxf03+5mGKwBJAlJCWihUadC2fVRpMJzASu108iCkQVVBXNx9TaU0m+7KfI2qmpBlkMaCtb2Mef9u8+Mg0N81gCyDsrzXnc4MS07sbHqsfnymv3hqvO+bR9fNcyjuca+H4ft6jX+zsUoZzAcaBhA6FiYGddOjzAyiu+5B7y+He09R6GvfL1wXLOvtjy1L/VncLxxH62693bFVBWl6/7G2rce/3bFS6u/aWxn7lkK9G2o0Gs2goiON0rzj54d/+PT4ILh7zOHP937v6bHz8/cf+9ATmfr0G3sqzgr12vZIzS8V9x37+ONSSSlVkpeqqip15eHyvmO7K4X69Bt7KskKFWeFev8HqvuOrbdK9QtfvKVe2xopKaX63u+9//26XnU0rqoq9Z0fy+47FpSqqkpFaa6+futAPfrhwQPHfuJru+rrt3pqbX+kfujPRA8c+ytf2FMff35L/eYrO+pP/IXJA8f+s/+wqX7qV15QP/avn1Xf8Z/tPHDsT/9/b6t/9qmr6qd+5QX1w3/54IFjP/b3vqb+85/9jPq//7uvqx/9ew++h//6H15V/82/+rL6v/6LL6qP/lc3Hzj2H/6v++oTL26pT722o/7Ofz984Nj3/9WX1V/6Z59Tf/x//m31+H/1wgPHfvD/8rr6wf/pt9Qz/49PqMt/6SsPHPv4j7yhPvaPP6n+ws99Vv2Nf3TjgWP/zN/YVf/qM1fV3/u3X1N/5adfe+DYH/xLO+pP/c+/pT7wD/+TeuS//Z0Hjn3PH9tU/8XPfVb9zf/Pl9X/7Z+98sCxS9+xrv7zn/20+pv/8gvqr//cVx84tvHUuvrL/9vn1Ceev63+x//4+gPHPvqRgfofP/GK+uVn19S/+PTVB4595rsT9Rsvb6svXdtT/+DfPa8s5/7z83u+R6phlKmqqpSUUs3Py/uOfeI9hSqKQn3865vqf/3k66q7fP814vzlXP37r26on/mtN9TPfvJ1tXAuue/Y7nKh/tXnbqjfeHlbvb49Vs88c/97mJ8/vaY9aI0IAnlq7A//8P3HglKffmNPvbo1VINJon7oT+QPHDudHl/3r/yVB193b+947I/+6IPH3rx5PPbv/t0Hj33ppeOxP/VTDx77pS8dr9f/z5++/xoMSv32bx9f92d+5sHX/fjHj8f+/M8/eOwv/uLx2F/8xQeP/fmfPx778Y8/eOzP/Mzx2N/+7QeP/cf/+Hjsl7/84LE/9VPHY1966cFj/+7f1eMO9+/RaKQeFO/oys4f9DCE0JTvGaA3eABmOi8lYBwxpQ7L3ve7brfuHvWvs6LNsR3j6RCCGXgweEtl3L1pxvki4NpexDCx0OTje8dvvLxLL8q0aaLsPPC6Tc/m5W0tGjlJHzw2zSVLHfOI6v6gsC0DcljrR0fmqPeLCx2fzpzB/jg7atvdL1Y7Pme6Atc2eH5jCNTuO7YTOlSmQCqBvO8oHZujFMc0mWYS602q3ec6AYEzYXskScvqgWMbvsW5louFIHcffOGWb9P0LAyhwdIPivmaSykFC4FNZ6n+JmM9ziw1yKoJhvfgpSctK2xTsFh3cGzvgWO7dYdnzrf4o48tsb4t+X8/cDSMsoLtUcKLO5MHjrNMA6kEjmOx7D74e2YZgnMdDaR/ZKmOYdx/LimgGTh3PHLv8B2TGwdTap7JRy7NP/DvYZsGndA+EvVs+Db79xl7mNpo7E8ID1SpUXf8/s7/n/iNAqXUW24JHSqyX5ivUff/YLGHTuJn3u1cfftDKKXuP9P+kMR4PKbZbLK1NaLRaNz1+2+kjTWd6pbV3iSlW/e4vBByfT9id5zQCR0+cLlxZJEQx3qxOIy0qHhurc8kqWgEJt/9WOcIVPzCzelMV8LjSvcYa6OUIq8kvq/4yq0+vWlOYFjUHJdpXtCte6x2NKbn+n7E3iTlwuIxXieOFXFWcW1vymfe2GdnnNIObRbrHsstn4fPBJxre3zueo83NjUr5EMXOrznbJMbBzEbMxFDyxD085hpVmEAGwcZZSV4qAY/+he/C4BP/tYLdJfbbI+yWYtOi/aNI4mJICskC3WXh7sNnrnQ4sWtEYO4ZHlO+yA1fJssg2lSgVKsD2I2+gmbg5hJVvJwt85DKz7PrvX5zVd3eGVjiqFMnj7f5r/7E4+xNcr4D89vcm17jDBN/sKHFrmy1ODTV/fY7mVUlQbxLtY9PnhpjsvzIb/56j5rvSkXFn0W6w5fuHnAs9fHbA0TipkBpyn0ZjLnW1xeavDUap39OOdMw2GaSNb2M9JSHun/DNOc270YBSx1HP7sB86w2HCZxpLdYc5mf8qXbvRYG6SA4MmVGn/zey7jeAa/9somX1sfsjNIkaXAc0wMw6DmmASWRaG0h5Trgeca7IwSigK6YcAHVjuklTZFzSpJWmjsSFRkTLOchabLuVbIfBBSlJrxM04LKqlYanjYlolpKW70RoyzisCy+ZOPr7BYd/nyzT6v7U1peRa3Bwm3+wmuo4irklJCwzE432qyM8lwLZNL8wHP3urx8taUvJKcbVs8db5B3fdo+Q5N2+HT13oIDBbrNlmpuD2IyfOKVmhwdsHnR95/hn5c8usv9IiyEt8WGMLAMAShDV++0aefpXRbPh+80KIXlaSx4D3nmliG4CDKWW74/ODjXXanGS9uDBG24oPnWwhDcHUzRVYKzzVOzaOslJiGQliSFzZHmt2T61baKC3oBBZfvNbn6kHEY4t1/vwHz9BpaSq3lJJnr03oTVOt3uwaLDZ9zQqyTQopeWm3zzgpafgWTyy0We8nbAxiTZd2Nbvqw5fm8B0T0zlm/1S5dmxX0yliSRsjX3vxBivLbX75+S2GecLZdsCfffoMqrJOta+VOr1uvffiMY37xVtTdsbpUSv8l7+2xVfX+wSOxcce6fJH3zt/tMmnqW5d3BmH159Ux3jBLBP3HHsYd7axiuL+asFvreWl4622pt7KWKWO78n3xVFr6s2u+24b6+6xb7WNdbh/j0b33r8P493KzokIw9O96QeNe6thuZJJlVKIkkmVIk2XSZVSGhUpxSmqeBCcfm6gDC4seqcUSEGfGN5zsUZW3gtYp3vTUsoj8GBoWVxYEKzMuzyxUjtKrt5bq5EV/tG5UinFxvhYdbnZMEgwWJ1z+dDFYy+sq7tT9icZzZrgO5Y7R1TVSZWirJJxmSJQpEVFWlTMhQ5/8v1dhDBYNEpCYgC+vndAPowoS0mn5miquwBDpHoDcWG16zCVCS/sVuxMM+KipJ6G2LPPzPMEnmehlKKXCZxE0awMlhyfTs3kUjdkueVytuXys5+6zjQv8X1JLYSnWgHTosXzOweMo4r/+JLkv7/c5kwr0IngNGW8V9KqOfiB4PKSx3zbQJouaVkSFQYX5kL2xilxlaMQ1CyDQVxQolAWjIqUWz0DzzW4NF9nmhdsTFKKKMOxLXwHDEcQlQZRUbHccpgLTM60fa5nY7biKZVVYnqgLK3DMiwSDvKEP3Kxi1LL9KY5liEYxlo7CCFRFti+QZlL9qMMvzJxMwPbNHFtwYcearFQ1xvucsPjubU+6SSjLCs6dYtMVbRDh6ZvMS4TTCVQdoUhFQ3X4v2XGry0Oeb5rSG7o5RJVuHaJr99Y4e2b/HFW32KXHG27TIuKnbSmHqlpRJcx2J1PuSPPTnPl24NiPIKw6mYa1mIgwqjlETSAEOLPcZ5xfXhhEzluJbBfNtjoeZSrwte3RoxLEqy/YhPX+vx8GKd95wPQaBFLFs+tim4ujdlvmny7PqYhmewO8lpeBZnWi4feVgfOPYmGVLB9dGYhZrD46s1RnEOAs2UNAumRYGUNpMKgjBgvT9lo68p8J5jaOaNZXB2QSdDxkzf6UsbBzQbgt004iubPS5lNa50Q17aHPOF9V2maUkzsPnI0jzvXdWJRVZKAlOwGB+Dfv0AJvsp2BWer+g0tErxXOsQAKwPQ6BObPgCZnPuqfMBY0MRBjAqZiaZnD6ggcbqnVy3svKYxj0qjx93XIdaDdoNE88xON91QB0LBXr3LcoJ3hueXsPuP/bucBzF2uCtqQW7rv55a9d96ziQO8cqdYeCcVjjsAL2dq5r28eJxDdzrGUdJz7fzLGm+db3xLcz1jDe3l77ZvFusvMtjnt5snwjfiMnTw4PQvXns8W3W/fYG2tQsmGI0zo8HOtkLNY1UHp3kpLk+nh1eSHkw5fmeKgb4s10fdJC+9yEjsn+tKIfZVzbi1idUUbzUtKtu5im4LXtCZ5tMldz+djDXQzTxE6Oy2INx2SzLNkb5ViWwcceaeCYgt96LWNSKS52AtqhwyvbE9b7MdO8YqHm8Nz6gH5c8KELbS7MhziWwSQt2Z1ogLemlerkaa0f89W1AWlRcq7tszvNWWr6bI0SXticMJymCKUQKAbTnC/d6JOWirNtn3FakFiaEaQA04BJWrI1jMmKit44YWuSYxkG3/fIAsM459m1PoO0QsoK35E4puJWL8IyTCZJTmBbvLI1puaYrLY99qcFg6jg0cU6Z9o+l7oht/s5r2xvc3sQczDJ6YQWD3cDxklBnFe0fJuXbg+ZxgULdZfFhkdSVLqSEOcMZ60BIQQt3+IgLrAsg1aoXckX6h5PrDSoEORlxXPrfb50s8fBNKfpWTi2QVZIqkriGALH1uaeoWcfvf7OJGOUFmR5xTgtMIRASsVXb/VJK8neKMOxBIXUVgOuaRJ6Fhc6PoFvc7bpk5Za72d/kvHQfJuWb7PW19/VhZqLY2qpAgVUleLCXEArcPjBx5e1PEO5yyubkjgvsUztBzdJS7p1l2Zgc34uIPC0j9urO1OiUrHUchECokyy0vKpORYvbk6wTHHE0BqnJbcOIoZxgWcbbPRjru5H5FWJKQRLzYDQsZBVxZdu9NgcakPIR5fqVDMpgrWDmL1JdsSovDKvWYmGgqKSs6qsw8tbIzYHKcMk5ym3yTTTLuEn5+WVbni0BgBHysUPdWuca/sYD7BXuNINyYuKwzzi+n7EdpkwF2pT0sMK0uG6clgVckxxz3XqrvXMt3n/+Tbdust8zcWxDb6yNrivkvGd69vvlpn0TnTZ/r2+p5N7wbuaOm8e7yY73+K4V8JyqAnxdvuHd50cHnSamenvKKWIkpL9GXX6UIcH7p6c5zr+qYX0/FxwF2XUtQy6NZeb+xGbw4RKagbJzZ7HhbmAj17qsN5PuD2Iafo6+bBmSVbTNRC2jXrmGSqp+KGnVvj41SGDWEv1j5OCUmkPKc8yWe0EPHWuycE054XNlDgrQSmkgp1RzOeuaz8bpcSRh0/Dt2aaGT55KfnyrT4vbIxY78fEuVav9WyD13an3OpFICsWmz6+UzBfc7EMg/1pgmUIvuNCBxNFoQTPrOo23bW9CYMoZ2+SsjfJUZVkruHSDm3mAhuhJElRIVB0PRuEbiWUSrE5zBBCewhJqdtL+5OMjVHGKJW8/+Ic7z3X5tr+JlvDlKu7U4QBEsUPPrrAXOhRVJJXd8Z86o0DAnfIpfmApaZL3bcZxQVKCUqp8G2TYmYmKQyDtKz40PkOH74yR1woFhtahG1zEHMQFWRlhRQglURKg2FckFWKl7enND0b3zVZCG3Onm2RFBXjJMdA0Yt1NaThGjiWYHOQMU51ZaHu2rRCC6UEl7sBz6y26DY8nl0f0Y9zWrHF9f2JdmeXuh36XVfmj3R95mseoG0fdiYZgWPSnWHczrU9enFJUQkars0HLzS5tFBjGBd89faQtV7MpfmAP/30Gc7P6eOhaVo8suiSlToB3R5lXJw3GUwzBIJuw2Op4VPKVFPYBYxjbTC6OUyYJAXNwOZMK8Sx4NWdKbd6sTZaFYKNfoxpGniOyc44O6Ibn+v4nJ8PkCgEGofTPjEXF+o2hiHJq5I0lyilHrhp3s/j6Uo3ZJKW7IwTJmnFoSv3aDDiA+99Gscy2ItyJso+5fwN3FMT5mSS9SC69sOL9SOF9s/f6BPn5T2VjL+Z8U5UC/69vKe3sxe8GzreTXZ+D+JeJ5j1fsLOOKEdODyxcozbORknv9Ddmstiwz21kC023Ltk3E9m+4cqoQD9uMAwNNDZc05YStRm1Zia+0BDvpPv5fxcwO1BTFZWxHnJOCnYGmcMopwLc6HWupgLONvy2RgmVJXihc3R0QKaf/6LuJbBxaLi6UybTCIEiw0fy4DXtsckRUmFBsA+dbbFwVS3MDzLAKEVhDcGCVmp9V9Wmv6Rh0/ds458vZK8YpqXTNICy9CAYtcyuXUwZW+c4loGK02Px5ebdBsurmvTrUss02A+tBglkvVBxI2DhFGScbuXsDHQdO5pWmJbBkUhiQvJIJdgmMyHNqaAwDZxHIvQNXEsi6fPNbi5H7M/zWn4Nr5jUvNtwrikE5gYaIuE0LZJiwmuBcOkxFCw1o9nei8V47QkqyTMXOIVgrprYyCopKRruFiWoBM4OI7Bsu1hmgIMgSEsnlkNuT5TAC4qRdvXiUXLl3iOiW0e+q9pAUbbFjOT2SZPnWnw1bURn7/VpzfJsAyTlaaNbVkkWc4orcgltH2bD6y2eHipTlRoYcCnz8/RjzLSvGL9YMLOMGGSaSHNnXFBXmkc00NLbZq+wzOrLXYnObuTlA/YWs04LSSbo5RWoF22tYmtwcceXuJ9qy1e3Z7wlVs9dkcJgzil4dv85Q+v8uRKk4W6y0Lo8LXbQxBgGIKFukfglDopmW3Ml4uAT71e8truBN8ycGyDbs3BNOBcO8A0BXXPYZoVLDc8JrlF2zOxLROBBtvXHG21sFj3EMD+NEcqQc01qLnaLdsawOPLdVq+xavbY7JK0YsyFDxw0zxcTw4tXQ49nvJS0o8z0lxScw1Cx9KKzsrmK//213jmfJv5fkw6SI7A/YfV2rejCXPnenakx7UTsTdOKSqJZYi7lIy/mfFOVAv+Vt/TybX9nVjZeqfHu8nOtyEOlUdf35lQSIVA8MSZxl1tqpPO5hv9mG7dnbmba2zA8xvDU6qcUkpe2hyzP0k52wl4qFvDsQwWGx5bw5Q0r065EgPHpIrD/76F0rJrm6x2dPsosE1e3xmzPdYoOXV4DcfikeUGyy2f5zeGM1O95JSbsD451nBsk07g8vhKnbyUPHkmZZwWmIY2dHx8WT8+zQoWGx4t3+KljTE3Z2JqFzohdd868vA5/NwmaUXgmLz3TAPfNvAtg0cXG+xOUw4mBZ3QxrdMAs8GIXjP2TZXFgI+/cYBr+9NQPnsjmNtBionZIX2NjJNCC0DQ9g0fZNHuzX245LAtpiv+VShZLHhYAsT1zFpeBZPnWmwO87I24pW6OBbBq7j8AMPB7y6N+Ha3pRPXT3ANCzOz3uM0pCDSUYpS2zTYHOUUUlFYJssN1wMIVhu+fyxxxcJXYusgt404/r+lG7dwTYM0koyTSviokKh6McFn7m2x1o/ZL0XoWTF9jin6Vt89MoCl+Z8XtgaY5sGz5w3afgWaz39GQ/igt98bY/n1oYaA7M7wTG1h9tK28NC8MJGRqnQQoqepc1caz4rjhbIfHQp5Cs3h5gGxIWCWFfTbENX8tZ7MUoIlpo+584ENEOXZuhyfj7EMcWR19I0LZkPbT54vkXNNam7NoWUbA4z3nOmzuc7ATf3I6Zpxddm3kyPLdfp1j1sUyAEmAgWGy5Nz+R95xo0XJtm6GIYBkqV2s9tmlP3bB5rBzyx3NRzzzOZr/k8tlzj+r7GwQS2wQsbI57fHPP4Yp2PXOrgOdaRRQAcJy9t36EfZ4zTks1BwkLdpdvwGUQFG0NtcotSb2nTPFlJaPn2kXllzTWouw6TLL/LsfwYoHf3dZTSmjCHSdqDKhN3tlAOzX/rno1n6+cP0+LIK+vtMLTeavxeqQW/nXbRt+qe7tWe/EarSH/Y2mDvJjvfhjgy+qwkAkEvSsnK2hH49+QXultzte+N0t5MDc/i0aU6r+1MmKQVQqRHi+LLm2N++eubpHnFk2ca5KVkmBS0PJuFunuXf1VWSvYmmX78hCrzm8XJE4wzM93cGCZHFhQnxx1KskNCYFtsj2KyUq+0hx41Jxd119YU3q+uDdgvMm4daPuCQZxrawoFr+9OySrFQ90aczWXx5ZqRLk8Ule+01n5bCfg4kKdduAwSXOmRYVjVay2A5ZmnmKHUvn6dJyzOdCu36ZpEKUlZxse86FBLy5wLINu3aZb81hsBtQ9k7WDlHGW0w5tVpoe7dAhrxTrg0QLJgrBziQHDN5ztsarOxNe2Bqx2tZqvlvDnCRX3OhN+dCFOcpKoQzBXOgwV/dYabmM45KoqFioezy62KBTdwk9m6KCumew2m4xFzqsD2LOdAKWGi5fvjWgjc0ozrh+EGMZght7EdO84HY/pqgUy02PlaZPhSB0HQopeebCHFe6IWu9mKv7U750s8/eKJ25ZFu0ApuskKzOBXzsoQW2hhGfvnaAKXRCsNxyqXs2k7TAsU0sEz752j5fvNGjlBWLDY1fe2K5zvtWW3z5Ro/n1sf4tqAV2Jxr+3dh01Y7PjcPpqz3Ez577YClhmYiRmnFNNcHiGlaYpgGjy41WO9H5CV8/kafSioeXYaykkzSglFaIAaK0TTnU2/s0wpc/tiTSzzUrfFbr+3z7PqANJcsNFyePtfm4SVtMDvNKuqepQUluyFZUXGzN+X2IMExBXFRUkgIDANXnK6yHreeDIoqZpIUpIUkykueWGmwOdJeUp+51uOPPr54ai7db2M61/FZ7fg4lsG1vQghdOIziHOmmTzVrjpMSO5cB+6cz2/m46SU4o3dCRuDhLNtn4cX66dc0A/XrqzUrupfmZl9/n5stbxT2kX38/d6O1Wkk98heLB6/h/EeDfZ+TaEEILHl+us92Ju9SOyQgMC74WhQWgNj05NS9ovNXzmay5LjZKTrryHz03yiijXi+juOCUrtQNxy3cYkp/ynXm7PeY7QYxHPfulOufnT0u0n5xYh5vCC5sD9nYG/I3/9j/DtUys115B2OFdCdJqJ2BjoFtQG8OErWHK3iRjoeYwX7NZ62kj0Y9cmv//t/fuMZKc1d3/p6q6u/re0z0zPZedy673gr2+37Ph/RGIV9iOw48E8r4RsoIhAZTEKIAjEqwEDFIkWyEiBIdAFJQQCSlGRjJJDPgXvzaxQ2J8WXvxete3tdczs7Nz2+np6Wt1dVc9vz+qq7a7p3suu7M7l30+0sqevlQ/VU9VPafO+Z5z2D8Q4/hcsenG2+pOBjxNwZuzBXRN5ZLeMNcMd3HFDieTzBOAY1OzoVytkS1ZhHWN0VSUvf1RogGn2mwkECLo95FOBNnVHWFnT4T37FMoGDWmFg3mio7n6pKeEJlSlVTYz1unnfAVwFBXhMMTi5RNm2KlxmAizJ60Qc6osbsnynXDMaayZSYWyoDgf13SzXB3mNNFk7l8BbPeyFPTfBRNm6mcwUA8iCWcMJ1lCcYzJUZSYUZTEd6aK6CpGl0hHwXDombb2LaonxuCbKmGz6ewo8tZrFJhncsGotSEcxz74zo+4L/fyTCVcTRSCd1Hf69OfyLEYqXGdL4KKKTCOlftiHP9zm5mChXmCiaxkJ/js0XemM4xlavgV+HKHYl6tpLKqazB6WIVW9jkKwrVmsWxySzZco3h7gj7+pz6PeOZMrN5k+mck7t6Yr7EJXXRcsivkQwHyBQr6H4/O1IwkAxxYq5Q91oKSpUqR6fyvD1XRFUEuYrCW6eLCBTSMUeUHNE1TswXCPo1ogEf79/fxxVDCRRFWbI4VGo2r5xaZCJTZK5QQddUIn4/UV1ru0i657mr2Xtn3qn91BcP0hd1BNblmsVbp4v8klHzavG02xYsXaz2pCP0xXWiusZbcyXv/hAXVZRd+9CB/iefhZBvyfXe6JHo1A/LvaaFbfPi2AKT2TKzOcOrqt66+CqK8HplbdVQy2YJF3Xq77XasbSeQ8P1ZJSN3q8LiTR2NoiqDbGQr6lbeesJrQCz+QqVmmh6QmsXG9Z9KsPJMFfuiFOzBTfuTKKqKjM5g3LFYnIhR82Goa4zOaZriTG7F0s7EWPrk6AQgjem8563Z7Q7zEze4OhkHjNXIjo92Xb7jU/yQ10hR+Ss+5i0nYaW8wWDiUyRsYUSO1NhCpUqeaPW9sbbeCMQQjA+X+LQWIbJbJmcUWWoK8yi4TyFu9koRtVpv3HTziRVy2IiUyLgU9H9jufHp0JPTEcIgV9TsWyYK5rs7I2iaRqJiEY8HGBn3SB8c6bgHYO9fVFGU2EE4FfhyOQii2UTv6ZQqdX41UvTmDVB0azyf1+b59RiiV09YcIBjbczJY7OFrh2qIubdqWYzDpP79Ggr258+dD9CslQgLFMiYlFg2TIx8yiYxQOp0KUTQtFgbBuMRwPMFuoUDQs8maVvniATNHk+IyzaM8sljm1WCYW1DBrTpPIXeko14wk+P9emeXUYhm/qmDUBMdmilw35CMS0NjRFcKybX5pdw+xUAC/XyVfrjGbMzxjZDBhkQj5ec/eXga7Qrw0scCxyTwLJZOukJ++eIBXJrP8+8vTBHwq/2tPD6OpsNclOxn2U7VC5A0n3V4oClcMJhjpDhML+nhrzvFUJEMBFGwerViUqzX6oiEyRZOpRQNVESSCfkChPx6kWLUYSTnXxfGZAuWKjV9V2NsT48ZR5zpq1cf0xc803bVR6I3q7O2LMZQMUbUBu/Mi6XgwNRQ3diycDvDv6ovx+kye3d0Rr3cdtF9wgebXUqGmDuQjqRDDKcfTqpRKMDaGAuxJRxjWQ2sOXTQulol6qQc39u1GxFoX380oIl4rm2UfzlUPtOQc8pJRtu7crBVp7GwQbraUUq+i7J7A7tOZG5JpPCEbxcjtrPrR7jDDKScVVa8v4Mmwn5cnspxadArSTS4a7OyNet6glVLYXdo1tnNet8iWq02uUKNq8eL4ApNZg9ncmSyvYEDDFzhzyhlVC71eRbE1fCcQTC6UeblYQQFSIT8LZZPpnJM5E9V9dEcaL9LmG2/r2E9my4zNl5jMlgj5NQpGlYjfh08RZIsVpnNO37F0VK8fx2HeniuRKVXojgS5pDfMofEsyUgAv6owmAiRNaokQwH8Kp6h5M5L2axxaCzDeKbETCrMSHeYUH3fjarTzPF00WAmZ/D6TIm9ON/NlWtMZh2d0GLZJBkOML1oULMhESzQHfYzPl9i1q9y3UiSq3ckiOgar04XmMkZ5I0ql3RHsGyLd06XGM+WSQT9HLwsTXckQMm0yJSrBP0W+3fEyJdrTphJUXl+PEOmUMW0aoQDPvb0xNB8CgN1wepoqouBrhCZsknNsqhZEFAVwMbv07h0IMpIKsKl/TFenswxs2hgWhZ5w2I4GWJff5Rfv3qAaNBPKOAsmHnD4vDJRSzbojvi51S2wjuZIoZpEdZ9GFVHgB5sENMf6A1TNgX5ikkqoiNswQtjC55hPZwKoQA/O34agSBfcXpLgSAcUFEVDQEYNZtd3SHeszfNnnSEI1N5pwhnQGUgEcRC8NpMkSt2JBp0LWVPL9cXD3LNcIJk2I+wBX6/43lt1em0W0xaQ0oj3WFGe8IIhSXVzDstuI2vCWjS953MnsmEaqXVuF/Nwtm4WAohuHJHF32JypLQdSObUUS8VjbTPpyLHmjJObSKZJTthjR2NohONXTenC00hWRWc0J2cnO7GUmWDTvqHp2hrpDndWiN17o3wHZdituJGFNh3RNFNj69ut8SCKfuDXD5jjgoMDed8bb5wjvzJHps9vRGmp46+uI6k1mDqcUypxYNBhJBdqeD+HwqC8Uamqbwy7t72D/ohKyuG0mSji/VDDWOfUc95dynKagKVC2Llycz/GJigbxZQwi4pDdST5ktkY4HuWwgStEMe4ZnOqZzMuOIUke7w3SFAswXDZ54tex5urxjaducqofggn4NYdsYVYuApmBbFtlSlZlcBQUVy7ZIhgJEdA0n2SwMokhPxE886Cfo0yiYNcI+jceOzTCxUKI/HqRmC4ZSEdIxnYWSSdm0qFqCnd0hEiEf/3FshhNzRUIBlV09IUwLTmUNCmYVFUd/Eqkv1Lpfoyvop1iuMV+ooaBwumRw9VCSVL2CNoCmOdl44/MlCpUqNcviZ8fnWSjVGEqFiQd9TOUqaKqChcCyhaOTqVSJ6n56YkEv87BSswn6FFAEuYqF7tPoj/vwqyplLAYSIW7e2XVmThVACCYyBrpfIar7sW2bf3t5mpJZoy/mZO6NpCLs7g1Ts2wms04D2oJR4139UW7cmWIuX+HoVJ6qJVA0jb39MRLhAP1xZyE/nffxzkKZZFAwvVhmTzpKqB6m6YvrDYJ7g2Q4gE9TqbmNQBqE/stdu+28uHMF0/EWFkx21uwmT1C7bbWGavtiTjmBvFGlXD2TCaUL4V2Tx2cL7A6Hm1LWV6PXaB3vnnSEvStoe9yxn0t4ZDOIaC+UEPp80ukc2ur7tRaksbOBtF5ElarVNiSz0gm5nJs7b1jEghq3X9KP7tc8YXLr553U0Twv1rNXrhtNevoX94bj1t1oLDx2fFZt0g6BYyhdO9KFWXNqhoxlSryrP84VOxJUomeecF89lae64Iiv0zEnRd4tvDjUFWKmXpQsHdPZl44hEKiKgqi7/R1dQHvNUOtx3tkT4codCTRF4fhcjoXTVd7JGNQsi3QsSMCvUrPCCCF4e67IazN5xudLXpaXo7M4oyU6Ve/cnitXPc2MU1AxQDzk5625EgWzhl+FhK7x5lyRbMkkb9Q4nTc4MV8iFvSTCvvoi4XIV0wiwRDJsI4Q4NOixILObw8ng+QrNY5MLPJOpoyCgqY650TJrDGTE54hlilWAKe+TlT3YwGaqjKZNRhIhPBpTlq4ZVlEfAEmFiv0RHUGEk6ROXf+EiENVJWYrpGK6OzudRbIuXyFqmXXi8gFGJ8vMrloUDRrLJZMusIBZnIV5vIGqVAAfxTm/CaW7YhaY0G/p3dBCPoSIYaSEbqCVfq6dBK6n7liFd2nEg5oWPVl2hXT54waU4tlFKBqC6K6D8O0KJRrBDSVhaJZz0DU0VQFv6YihKBsWlg2hAM+dnb7qFQFuUqNy/vjxII+7/zuizthyqrlVJ6eqouy3WvBFdw3CoHzRs3zdjYK/d3ru50XpZ2ubDXp5su9trcvynAqxHjGKWjo6fmKtldUcDZvkK4X4FyLXqPdYtlJ27NebBZx8NmyGQy1RraD0XYuSGNnE3EmBNM5JNOOTm7udEynUrVIhp0F2H2ibvd5N9TjNMsU9eagZ9pDtN5w3Btdu6cFJ+QR5rkTGWZyBi+NZ9nZHVliuJ2YLxLs0pgvGtx8SY/3e64BM9IddpqFNdQVObVYoWxaTU+/7RYVoOlGo/s1+hIhXp/JkwgHKS+WiekaZVOhXLVIBP1EAip+TSVbrqIKOD6XZyQVbqo5MpwMe7oIFJjRFMJ+jdMFg7F5p87IvnSU+WIFDchWakznDV4aW6BUtXhrNo8AAppKf1ynOxpkJl8mazg9o5wwoJNZdtlAnJ6okw4d0v0M99S4smhyaX+Mq4ad0MrhiUUKRo3eSICorjFWtZnLm5wuVTl4WZpYMEAkoNATDaKpUDBqjC0UKVUsKjWLctVGFCpMLpSxURjtDtGfCFE0HbH1O/Nl9ICPQsXyUot1n4JhWrx4MotZsUmGA6RjQa4fTeHXFMYWSqiKys6eELvTUY5O5jhyKsf4QpneWBGzLuwFuGY4wQevHmBq0WC0O0x/TEdRFA5PZhHCKbrozqPrWfRrCmOZEkGfRtCncml/DCEcL5JbPDOqa/QnHONxOGWTCvtJhQNYNiTCfj72y6OgKHWdT6kphDrYFaFmQyzrpyviX2LAtBb1AzqmbLfq1xrLPrQzVs4lrKAoCqGAzyvyp/scQ88wa56xk15jFffW7a9FEHuuC/1mEQefDVvdUNuOSGNnE6H7VK/4mVvGfTV0CokJWziZTItO9opbvLCTwHmoK8RsPdNlKBlasXjVsjc0RakXpzvTf6eVWFBlrlihUHHCO1W78euKp+uo1Jw3Go0N9ybdmCHmuuZdQ8R9unVF1Jf1RxmfL2ELwUhXiKFkiEzJyZDKlquUa5AM+7g50c1r04vMFZwidm4Gm3vcXE1IwKd6WV5PvXmaycUyh8cXODaVIxLQqFiCZMjHfMF5+ldVR9RcrNTY2xthMB5itlAhV6mRUJ2F0u9TOLng6EJem87TH695N8p9fTFP5Nxo/JUqNV45tcipRYPZvMFIKsyORBBFVbikN0xXyE9E1xhfMNA0Z16KpknRsAjpGuGAxitTeU4uGrwrHeX/3DDE6YKJgkJVCJLhQMMC6XRFn14so/s0Qn6NKwfjXD+aJBr088LYAgFNxbRsSqbjJavZoAEFs0apUmViwebkQhlFwWl3EQvWC/I59Zl+eU83fp+GX1MYSrVUHu8O41fh1akCmVLFa3NiVC1eGFuoH2enXcrlg3EQMJV1Wo2UKlUKlRp701FH04Zj/LUW6tyTjrT1kLQ751dK2W7VrzVqt9pdx2e7mLeOS/epGGaNp984zdjJOT5V/9yedBRFVZvO47VuezWfX4+FfqPEwRe7obZdkcbOJsHpAlx0Cu7F136DWBISqzmVZmdyBplihdNFZ1G/YkfCC/+0pny73pRGzU6nG85KN7SgX2uvpVEUxP79VKo2tnDGsVAwOHYqT9ZoFjq3+41W17/7vlusLW9YVKoWNdvp4C2E8AoZOhlMGkPJCKmIn6t2JJhYKLNQMumtCk93M5x0CqLFwwHCAZ/n3XGZaOhb5OijVHamwhw5uUimVMOn1Yh1hxlNhpguGEzMGwx3+dF8UDQEO5JOHZ6pfIVixSIR9vNLu7rZP+ik0ddsQaZYIW4HcOsoeR6shoXSzVo7MV+katlULYtkxM+l/TEGu0L8YnKRd+ZLlE2LrrCfrnprgKEup7eTALrDAYJ+lX8/Mo2GIzgfTASJ6AH0hmKPqtpcKyagqZyup9Jfko7REwtSqdn01UNhkYCPvFHlzdkCBaNKqVojoMDbp0ukwgH8mqOB6osFWSg7eqOXJrKczJZJx3R+67pBVFVFd7tnmzXPyFMUpwhnY82Q8QWnp9ZsvkI6qnN8tsDlg3Eu3xHHtGyOn5jHr6nsTAQwLZtHXpp0PhvT6Y7qbQt1NnpIOp2PK6VsN9Q2b/l7/bBtm2On8mRKFS9D8vhskbdPF3junQy+iklmdA+JsB+toVJ763nc7n5zNobLWhb65QyLjRAHb3VDTdIZaexsEhqr/ipKdUnDTpflbg6N77memskFpyprKOCU2680CB9bcb0pra+1u+GsdEPrqKUJh1GOHsUyTAL/+RZmpuS0BMiVqVg0bavTb7i/05wO7HggwKRccRqUKorCaDLsGUFCCLojjkA2HdWZLZhky1W6I44Y2a23AzR5kFqNyJm8wWLZadg4kgoRDPjY2RPhmuFEvUFqhf5EiFREp1i1GEo67RaiukbE77Q5CPgcTYxPU7huNOl1jp/NV6jZ9Y7RfpWukN9LcW479wpoClhCMJurYNXHGdE1IgEfpmUTCmhoitN6IqBpXL4jzp7eiLMNRUHDJlOs8tbpApcPJIgG/fTGfc0eC6XZN9eY+ecWtHO9ajeOJhHAm7MFp52IUWVywSCm+6jYgtFUmL5EkF/alSIeDnB8tugV2TtxusgbswUUnPPOLWD33IkMPlXh+tEUe/uiTdWJc+UqMzmDkO5DyVfIlk3+6/hpwOnxVjSdKtRVW5AK645oOVNiatHAqDeuvaQ3wttzxaZCne655oZHhW0zseD0WGuc+3bz4nkcfarXKLM3FiSwzoueEIKjp3L81/HT+OuGjNtWpmTWiOkaWijOC4/+Fwf3p51W0qzeIDkbD8VqF/rVGBYXWmfSnHXWviXPathMWVwSB2nsbBJWc4NY7ubQ7r19/TGGUyHeqnuMGtNil6P15t3uhtM43k4l4Ze7UWmqylBXGEs4fa364iHPsxPQFC9zqZNXqVJ1jJd09Iyw2W2E+IuTWeK2U4RxdzrCyQXNE1G7IS2E4PmxhfriZjbV22kUY7feqHSf6omB8+Wq135D92uMdjvhkWQowFAyyMuTi0QCjjdlOBnmnfkiEwtlBuI66bhOKKBRrNiYNZu35koNJeBhT2+Eak2wUDI5Plv0ntabitpVLU4ulCmYFgXDQq3rfWZyBsem8hQqVXamIkSDjsh4JmdQrjp6p319MW+hfnOmzGWDMa4Z7iIR8vHixCJ9sSC7e8Oex6AvFnQ0SjkDw7TR/cqZqtOWaDI6zZrNTK5M0bSIBDQyRRhJhanULGK6j7GMs4jM5E0SdfFzIqQxFfbx9PF5dE3lyGSW2YLjpXlxfIFD4xkiAacSuGk5lcHditozeWdMqbCfQDrGWKaIX1WYyRns6Y14DXEjAR+hgMJC0UTXFIaTQQYTQSwBb885xT2jutrUpLOxvlS5YjFfrHgeobH5Env7okvmBVpKKdTnMlOqcHy22OS5XG24pNNnKzWbhZKJX1Wo2japsFNU0DBtpnMVLhuIc/OuFIlwoKn/3moNkrPxUKx2od+MoZ52JQYasyzb0WluLnZB8GZDGjubBEVZWmOnlcZeWWbNZjgV8jwxnW4cYd3PFUOJNd1Ul3vaam006mobnnsn0xTy6LRtz/Pk17huZ5J0IshQMsTedNTL3npzpsBs4YzeplWL9MZMnhfHnKyxa0e6uGE06YU3GrNlXK9MU4G1urhaCLHkJr7aJ83GrKzGsvtLsmviQcbmSyiKn7xRw6w5qfhCKPg0ld3pKG/PFSlUbC9N2N2GEKKuQXF6inWFfU3aknTMqb8zmzMQQrCrJ8KJ+QJG1WIubzrVsy1ByK96xSgDPpUXx5zeUu5C7XbJLpoCy29Tyjs6JSHK5I0qz76Twa+qmDUbRVHIG04dIEUAah6Bo41xj2VXyM+Rk1l+MbFItlzlqqE4u3ui9MQCnC6YRAIak4sGQb9aF+3q/PT10xyfyxML+EgEfbjBnpJZ42TW6RXlNNZ0wmpOBpSFWddyOYJup+hmVNc4dirPK5OLzOUrjGfKjoFb70f35myFaNBHtxUkHQuQjoUomFUKFbupSefxWdXLGpvOlZnJVZjJlekK+VEUxzCeLVToTwRXLPjXF9dZKJtL2rusNlyy3Hnp1usCSIYDnvEZDKgMxHV8mkoo4FtyTa7WIGmXNbaa+jyrWeg3Y6jH3d/GEgONnr5WpBB56yCNnQtMp6cAV7OzXCd0vV5Y7WSmRN6oNqXEruXGsdwT5UqC5HYX9kzO4PXpAlU7h0B4uqDW33xzpsDcbIYb/vdtBHwq+557jtFUN8J9f9YpjDeXd7J+gCU3mUrNEbdOZsuAQjpusLPnzA3GNRrTsYBXT8g1nBoLrLW72RtVi+lcmfmCiRCi4w2unVDa3WbjsTJrgmypStCn4ver7EiGCAU0eqMBbPuMN6GxUWNjZlnjE+arUznP82CYNofGFpjNV4jqPkIBjcsHoiyUTE5ly+xIhuiN6Rw5mSOnwkyuQiIcaDLSZvIGVUs4XbENi2jQyV5C4FQqrtcQ8qsKVcvyKjbP5iv41YjnPVkomZjWmcaVtm3z0niWomkigKoluGYkyd50xKvw6xgUJjOLZR47WuGxY9PkyzViIR8HL+2jOxKgKxzgjZk8Cgp9iSA7kiHS8RCXD8a8VgiuZ8edXzfcsDsdYXKxTLVeuK8/4eiCjLoC3hHOQ01AsVqrhzbNJt2Xu8AFNAXDtJnNlSkZjuGV0AN0RwP0xYJEdY2ukN+br3Zp5O2yn9bi1ahULSbq4ejWa6JtsoEi6IsHmVwoY1QrjE+cZu//e4tjQj7/PITD3nfdsaxk8Ljn5Hou7Js11NPuoanT/XQzeqck7ZHGzgVkuZvFcp3QXRTFKeh2MlumXLU6psS2agcaU1/bud1bwzSdjCZ3jI3GgO5zigtW7Zy3+LXTBbk3hWKphv7Ga4AjrBzPlB2PRbnGWKaIT1UJ+lSvc3LrTUb3qQwlG7LGuppDc+4xfnF8gZol0FSIh5weQ+10RY1/uwvbqcUyAU3Fv4xmqp3HqbEgY6Vmky2bhAIaVVs4HdXTEe9YvDy56NVAamwD0ji2xifMXLlG0K96TWAzdcF5KKBh2/DGbJH+eIgdyTBDXSFHlF2qoCkKU4sl+hPOousaaV0hP/NFgzdmilQti5t3dbM37ZwLo7WIV0MJoGBYFM0asaCfm3Y66eXHppwWD25o9MyCqHLjzi5mcmUKRpXRVJj9A1F8Ph/7+n2M9jjZa//zdobFkslEpogqFATQE9ZJhgPYwjmParYgGfET1n1Nx6jVAzpai3jnQNmseSUAfKqTwdaYSbanN+KEwUpVCkaNfemYpwFqVzeqUnNao/S6IdtYkHRc55rhJLGgz0kqKJmkwk4WlzuHrddiu+zH1epaxuZL3vm+Nx1dsQZPq/dxrmCw79gxd4NLrpXVGi/nY2HfrKGe1Rpim9E7JWmPNHYuIJ2K+bmp08lwgKot8KsqmVKlrdGgd/AqQPsbR2vqq6OVWD6VvJNexTUGpnIGkYCPgOboefYPOgX/Ghe/Vtybwmyx4L12bCpHBj/zBZNTi2WwBSdzJXZ3O1WB3cWj1RPVmoLdqmFw6wUJBH0xRxvT7kbUul3TEuh+BRWFsUyRY1P5Ji9Vp0rVhuk0kXxxfAHLhht3OQUZvfBCKMDudORMz66A8J4a01G94w3SfcLsiwWZXFggb8B8wWmsOLVosLsnyvXDCQ5NLJIr1wgFNC+k9+yJeQqGxULZJKA5x2ggEW4qDHnsVJ6qna+notec8If/TPuQVnf+bKHCaE8ETdOcApEddAq7eqJcM5Tg1Zk8pxbLvDpd8I6jaxANJ8P4NYVIwEdPVKdiCX5pV4rBrhAvTzr749dUwgGnsKJr6Lge0NYGm7Ztc/RUjpOZIq9NF6jaNgNxpyN4YyaZq9WKhfyE/Bqj3WFUVW1bN8qdczfDrCsU8LRKsaDP0yq5uq/GpIJ2Bshqr7NG3LYSsaCfUEBbkhnY6XxuqgmlnTm/3BYtirK08fBKxsvFtrCvxhDbrN4pyVKksXMBab1ZtLZt2D8QQ0HxUkjb3UzWenG1pr6ebSo54GkBGpuXBlVQVbXj4tdYB2ckFSKtJrz3siWTVI9jMIT9GtOLZWbyFSp1AaqbBdQ23bdDrZLWekHXjnQ1eb/a1eVxt6v7nIafQsnjV9UlXqp2XeknMmUmFkqczBR5dSpPqWohEOzsjnjdrcfmSxwazzaNv1Hv9PzYQpPx1CoOd715JbPGobFFMqUKAU1F9ytMZCvM5gyvHsxMvsLedJRkOEA0qBENhqjYFpliFVVtXtDaGamtBmAnd367xdv9TsCnYlowsWCQDPrJFJsN98ZzOKA5i67rEQO839ubjjLaHW5qXdKuoaEA3pha5OEXJzFMi6otGEgE8WlOtlnjeM9otZxQU8CnUjZrTeUWGkM207kyyVCAG0a6AHir3i/t+KzaIChf/eK/Vm/KmeuVJZmBy23XrRU0kgqhm4b3uUNjGXr7FO98X8v45cLens3qnZI0I42d80ynImSd4vaN9UPW42ai+zUn9TXuCIGD9R4/a00lh/bNS13aXfCtmSyapjCs23TV3++Lh9g9GMO0ogjb5r/fmse0hZciHNAUcuUq07kyOaM53bcTitK+XlDjeNwwjit0bdzf5bxUrYuDgiNELVctFBQCfkcL4lOdsIyiKF637tbj6hoyTh2kqhMWbOhc3bgQBuq/O71YIiMEIb9GTUAsGCBbNgkGNGzb8UrN5iuMpMIEfCrJsI6mQndEJ6Qv9W65RqobfmvnNVlpgWu3eJuWIBrU2JUKgwKpyFLvVaNRoSiKZ/S4wvfV9JRyM6LG5gs8fnSG12by6H6Nm0a7uHK4i0t6om0XoeGUU0sJ4I3pPC9NZIHmFilnwsoFqpbFWMaZt7l8pd6qpFlQvtrrda3elNUaGK0p025tqb5YkL3RM9/Jly3s/JnfXev45cIu2apIY+c8slIRsnZPVivdTNb6ZOiGfUZSYc/L0+43VvOUt5qFr9FQa+yU/vLkIrqmkAkrvKv+ebeaq5MdpTKSitQX6QCXDUQ9wXa5YpEvV5tSvYGORqGiLK0X5H6+MUU6FdZRFHOJx2JPOtq2GWq7zJRGLcjVwwnmCmZT9evljmujRijs17CFaBvmfHO2wPRiibJp41PBr/u4cqiLy/qjHJvKczJTQlWdtPN0VPeaSiajAUJ+jV++JIWiqk0GReN+ucXlXFGyZ3w1FjNs9OJULS+E2Ck0O5Bwwi2psM5lA9G2v+uGnjLFCpWGoo5uaKodjXMghOD5dzK8Np3n+OkiCoJE0Me793Rz7Wg3oUBzVqOrX5tYKGFZAp+mMJl1Cm8qCl6LlKBfa9KiqaiMZ0oMJJywZMi/VFC+Eo0exbNJ5V5LZlNXyN/coDeoe+0iYiGN3g4eOolkOyONnfPIagrvna8nw0bDA1aultoutNDJkFhtCqZ7A65ULRJBH1Vb0NQBbBlDojEDJehTSUYC1CzBbKHCSDXMRF3Y7Ka7u8ZVp+MohFNbp7UuT2NxurJZa2oR0Ji91Wn/W42f1jEsN8dnUoSDBAMqqrJ0IXSbw45lSpg1mysG4yTCAXb3Rnj7dJnZfAXLht29EcK63wv7pGM6Zs32vHlAWyO58XyybZuCYXUUaNu2zSuTixwez+LTzhT4a2ewN55Lb84WOLlQZigZamoue+xUnv86fhpVKNjYDHY19yHrhCvwrVSdbvGaqrC7J4IlbPb2xKjacHLBWDJ/rn5tbL6EadlcsSOOX1Xoi+v4NdVrkeL+huvlazTG+mLBJaG1lWgXYlrvUFDrMT8+W2wIPZ75jetHU+gJmR4tufiQxs55ZLXekrU8Wa1mm6031+FUaFkDqfFpXfepy2ZrdaKTEeZqU4aSISazBsO6QIyOOl6mDoZUawbKnpEuFEXxjBAFmFoscXQyBwrYwjFYZvJG21o/jccjHdO5YTSJ7lObDJ03ZwpMLDi/GQu1T3tvR+v8tft8pzleEhZsE1ZwTUOfqhAM+oiH/Awnw6iK4hTTq9r4NIWw7rzu6l68L4rl56fVIyCE0VagLYTg6GSOHxyaZGKhSH8s6HhCeiIdm8EG/RqGWePFsQUms2VmcwajqTDBgNNlPFOq4FdVqpbFSCrsibFX8nY0zmdvNMD/s7eXywcT6H7HACqawqtb1PRwUf+vpkLc7yMR9LNjIERfTHeqJbd4glRVZf9AjPmiTirspyaUszJQLlR6cuN51jQn5TKMjgI4hq80dCQXIdLYOY+cD0Hfajww7UScnQyk1iJ9VwwmvAJoa7kxdzLC3JDSvr4YA4mQky78zjvLbsvNQInqPvw+lZGU0zyxsZNzwbA4kSmRDPqZzZVBUThxutS21k/z8YDRVLjJoHONwbLp1DFpDFOcK2fT+6fVeHJ6jAUZTOgMJM5kJrkhtFYhr1F1MqeMqu0VPVxufhrPp2OnFIRSWCLQdo9h1bJAgN+nNnlyOgpnz+xt09+uoQd4oS63XcdK10lrcc3GMF3jvLbOX6N+bUdXkNFUmPFMmVem8qSj+hKPjW3bPPHqHMdPF9jTE+Xg/vRZXcMbkcXUNCfhMKxwzZ0N69EwUyK5UEhj5zxzPmLirhvfyxZpKUK45Oa6jBCxtUhfb0ynLx5EUaprujEvZ9g1FkxcTZVlhKA3EuBkpgQVmFgoO+0NGrKiokGNS7ojCAS98SDVmtMI06+pZIoVcuWqZxS0Hg8BHYzBpYbDubAevX9cwfVod5ix+RJHTuW8bXU63p20YJ0+3ziGxtBNMhzAtm2EcHqtDSfDXLkjwWUDca4e6lpSB6odnRrCuuNxO2+rqkpQW93x1n0txTXr58dKDxeufs01ml2jerFU5WSmxMlsmeFk2JunvFHj+OkC8wUTKHCz4bRdWCvbIYup1bCRlYMlWw1p7GxRlitCuBqPgYvuay7S597s3RDPWm5g7qJp2za5cpVY0ClT35TZsooqyzN5g66g0wfJqNlNxRPdMbsC2GQogN+nkC2ZjKYiRHSVSlUs6WnTXljc2Rh0iwSuRoi93BytR/hiuayuTmGz5UJLy6GqKpcPxjk6meOVk4u8MrnIdSNJ9vXHOma5rTT2tg1h66ym83a7bS5XXHMl47FVPG7WbPJGlXK9sax7bGNBH3t6ogiRZzQZJqqf/UPLej30bIQ3pTUM7CY7yMrBkq2ENHa2KHo9a6lTEcLV3ly9p92Uk92C4jb+PDtXu23b/N9js02u/8bMlmC1wujt78UK+ND+62mUeul6WJot1R8PeemzrWnurUXiChWbWMjnVRjOGTWqVslLVV9OWNxqDK614WqnRedswxftFrTlttXu8+eyuJp1IfjkotuSo8xoj2NMtMtyWwlPUNwyxrMR23vHo0NxzbUaA24a+vhC2dOEudtSVZVbLuulOxqgUKny1lxp3T0Ya20GelbelHIZ3vMe5/+ffhpCoTWN0QsblusesAXHS5eOnRH7X4jQnERyLkhjZ5Ox2pufoihcPhhfsQjhalAUBd2vrYtbupPr3w2PZOcyxF/5BQBvTufYsyvUcUFvzJbqFP5pbejZE9Xpi1edisMNqeprCR+1Kx4IjupkSQhsmSfaswlfdFrQWsM+K33+XGj19rW25FgrncZ4NmL71uPR6o076+aa6agX3mr8TtWGUtVqata6Xh6MC9auwbbhhRfO/P8aafKAlateA9wbRpNtj5lEshmRxs4mYq03P1VV160I4bk8ZTfiuv7B8ezEgj5vrFfsSJDTz0hWZ/MGwx0q63rC1xXWWEVp7pWkqmrHruSrpXERTkd1xuaLvDSeBZyKzI3p6ysZAWv1sKw0D61hn/PVr2i5lhxrpdMYV2MMthPbo5zJimptEttqpCpK+wyq1R638yku3irtGrx5qlcDdxuvnut5IZFcSKSxs4k4m4VrPbQA7WrQrOUpuxFVVTm4P83NRsozPhrHGq+ndQOk2/zOWvenU6+kTv3DVkNrmOy/35pnMmsAgnQ8yLt3d3fUoLhjOlsDdLkFrVMBv/OxACpK55Yca6XdGFv7OLm0Hrsmw7NeMbmxi30nYXY6qi+pmbSSiLvTcViNd+5s5nytc7eRQmf3fNjXH/Oy+6ShI9lKSGNnE7ERT27tatB0emJbrTGmqqqXtdK6CDRud0967SGXTlWa29X3OZdFoTFM1thrqzeqE/Cpy2aTrcU71+74tIqpXaH0WrOsLgSrWeTb7VO7Y9Tp2DVWTH5hbKFtI93WY+FqudajoOdKBvjZhhLPZu7WS+h8tmz070skZ4s0djYRG7FwLalB092+0zis3Rhrl8URFKKpbcVaWK5KcyevwbkeQzebaCQV4vhskYWyyZuzBWdf2hiFa/HOLadHaWxG2fj+2WZZLcfZHq+1LPKNYzTqWU+L5WpTv7Plwl3t9FmtjXQb27G0frbdubpeC/e5hBKl8SCRXBiksbPJWM+b32oWsU7GwmpFocvhtnwomTUvi2NEF+w9y/1ZjRcH2nsNzgVFUVBUlaxRJWfUmFxY4ORCc00Wl4Cm1CsRixUNwpUWSTdlf75gNvWqWi7ss1bOReDc2nyyL657tY2WQ/eppGP1OjkNIvKVjOnW82+543chHxzW2yMri/UtRR6TC892O+bS2NmmrHYRa7couE/enRaR1TY+dFs+VC0bn6o4gmHTZE9PD2dz6XRaVNp5Dda7/ke7jJTW7BxXP7RQMkmFdfakI8veJFZaJN1GoVM5g0jAR6Cl6N56ZGKtxivR6abnjl+IMoZpL6lt1AlFUTqKyFcyUBrnejXG0YXwmqynYXXeivX19Jz7NjYIWcDwwrMdj7k0ds4jG2kZr8W13roorMeTqludNhb0E/Q728saVXpjMZidPav+PI1ZIaLDZ86nYLddRko7AXHesFAU02n0uczPd1ok3fNGCKf55GAiRDCgLtneemRirXS8lrvpuePvi+v84mS2vt+rG0cnEflaDJS1GBnn+1rcDCGxjkQiMDd3zmPbKC5UbzHJGbbjMd82xs43v/lNvvrVrzI9Pc3VV1/Ngw8+yE033bRh49loy/hcFv31eFI98/tLO4yf63EYX6bq7nqMvdPCuFJGytkc89ZFslXn1BcLoihK2+2th2G30vFa6abnZtg5DU3Xtt/r4Q1ZjZHhHtN2rVU2GxuVXr6ZkcfkwrMdj7kihOj0kLxl+P73v89HP/pRvv3tb3PzzTfz9a9/nYcffpjXX3+ddDq94vdzuRyJRILFxUXi8fi6jMmoWjz/ToZcuUY85OPGnakLbhlvdMy1sZv6etXkON/H9VyN1HM95kv2bzTZVFdmvX9vJdqKzNvM5Uafa8thVC2eOzHvtVZ5z57eVfX12ig287HcKOQxufBslWO+2vV765trwNe+9jU++clP8vGPf5z9+/fz7W9/m3A4zD/+4z9u2Jhcyzge8m2YZew+9W7kiTqeKfPC2AJvzhScej7lMrz3vc6/cnnN2zvfx7XRkzGTN6jUVldx1u2lBedm2C3Zv7ogeaUsp/M1x64H5sbRJAia5/ICjuNc0H3tW6tsNtbrHFrCOV5zm4HNfH5tV7bbMd/yYSzTNDl06BD33nuv95qqqhw8eJBnnnmm7XcqlQqVSsX7O5fLrfu4Nrr+yWagbQjEtuGpp5wPnEXp+vN9XM/GfbueIcvNeN4oitKxEelWQFHWr7XK+eK8hr3P8ZqTSLYDW97YOX36NJZl0dfX1/R6X18fr732Wtvv3H///XzlK18572O72GtotDUczHPf7vk8rmdjbKy3mG8znjdbPYa/nq1VzgfbURAqkWwmtryxczbce++93HPPPd7fuVyO4eHhDRzR9mQzeilWw1qNja1uCKyGrTqXjWxGI9LlYjiHJJKNZMsbOz09PWiaxszMTNPrMzMz9Pf3t/2Oruvoun4hhnfRs5kXmPViOxgCq+FimMuN4mI5hySSjWLLPz4EAgGuv/56nnjiCe8127Z54oknOHDgwAaOTHIxsd3EfJILjzyHJJLzx5b37ADcc8893HXXXdxwww3cdNNNfP3rX6dYLPLxj398o4cmkUgkEolkg9kWxs5v//ZvMzc3x5e+9CWmp6e55ppreOyxx5aIliWbhHB4o0cgkVxcyGtOcpGzLYoKnivno6igRCKRSCSS88tFVVRQIpFIJBKJpBPS2JFIJBKJRLKtkcaO5MJiGHDHHc4/w9jo0Ugk2x95zUkk20OgLNlCWBb8+Mdn/l8ikZxf5DUnkUjPjkQikUgkku2NNHYkEolEIpFsa6SxI5FIJBKJZFsjjR2JRCKRSCTbGmnsSCQSiUQi2dbIbCzALSKdy+U2eCQXAcXimf/P5WR2iERyvpHXnGQb467bKzWDkMYOkM/nARgeHt7gkVxkDA5u9AgkkosLec1Jtin5fJ5EItHxfdkbC7Btm1OnThGLxVAUZd22m8vlGB4eZmJiQvbc2gLI+dpayPnaWsj52npshTkTQpDP5xkcHERVOytzpGcHUFWVoaGh87b9eDy+aU8UyVLkfG0t5HxtLeR8bT02+5wt59FxkQJliUQikUgk2xpp7EgkEolEItnWSGPnPKLrOvfddx+6rm/0UCSrQM7X1kLO19ZCztfWYzvNmRQoSyQSiUQi2dZIz45EIpFIJJJtjTR2JBKJRCKRbGuksSORSCQSiWRbI40diUQikUgk2xpp7JwnvvnNb7Jz506CwSA333wzzz333EYP6aLk6aef5gMf+ACDg4MoisIPf/jDpveFEHzpS19iYGCAUCjEwYMHefPNN5s+k8lkuPPOO4nH43R1dfF7v/d7FAqFC7gXFw/3338/N954I7FYjHQ6zW/8xm/w+uuvN33GMAzuvvtuuru7iUajfPjDH2ZmZqbpM+Pj49xxxx2Ew2HS6TSf//znqdVqF3JXLgq+9a1vcdVVV3lF5w4cOMBPfvIT7305V5ubBx54AEVR+OxnP+u9tl3nTBo754Hvf//73HPPPdx33328+OKLXH311dx6663Mzs5u9NAuOorFIldffTXf/OY3277/l3/5l3zjG9/g29/+Ns8++yyRSIRbb70VwzC8z9x5550cPXqUxx9/nEcffZSnn36aT33qUxdqFy4qnnrqKe6++25+/vOf8/jjj1OtVnn/+99PsaGZ5ec+9zn+/d//nYcffpinnnqKU6dO8aEPfch737Is7rjjDkzT5H/+53/453/+Z7773e/ypS99aSN2aVszNDTEAw88wKFDh3jhhRf41V/9VT74wQ9y9OhRQM7VZub555/n7//+77nqqquaXt+2cyYk685NN90k7r77bu9vy7LE4OCguP/++zdwVBJAPPLII97ftm2L/v5+8dWvftV7LZvNCl3Xxb/8y78IIYQ4duyYAMTzzz/vfeYnP/mJUBRFTE5OXrCxX6zMzs4KQDz11FNCCGd+/H6/ePjhh73PvPrqqwIQzzzzjBBCiB//+MdCVVUxPT3tfeZb3/qWiMfjolKpXNgduAhJJpPiO9/5jpyrTUw+nxd79+4Vjz/+uPiVX/kV8ZnPfEYIsb2vL+nZWWdM0+TQoUMcPHjQe01VVQ4ePMgzzzyzgSOTtHLixAmmp6eb5iqRSHDzzTd7c/XMM8/Q1dXFDTfc4H3m4MGDqKrKs88+e8HHfLGxuLgIQCqVAuDQoUNUq9WmObv00ksZGRlpmrMrr7ySvr4+7zO33noruVzO8zhI1h/LsnjooYcoFoscOHBAztUm5u677+aOO+5omhvY3teXbAS6zpw+fRrLsppOBIC+vm2tN0QAAAgdSURBVD5ee+21DRqVpB3T09MAbefKfW96epp0Ot30vs/nI5VKeZ+RnB9s2+azn/0s7373u7niiisAZz4CgQBdXV1Nn22ds3Zz6r4nWV+OHDnCgQMHMAyDaDTKI488wv79+zl8+LCcq03IQw89xIsvvsjzzz+/5L3tfH1JY0cikWxK7r77bl555RV+9rOfbfRQJMvwrne9i8OHD7O4uMgPfvAD7rrrLp566qmNHpakDRMTE3zmM5/h8ccfJxgMbvRwLigyjLXO9PT0oGnaEvX6zMwM/f39GzQqSTvc+Vhurvr7+5cIy2u1GplMRs7neeTTn/40jz76KD/96U8ZGhryXu/v78c0TbLZbNPnW+es3Zy670nWl0AgwJ49e7j++uu5//77ufrqq/mbv/kbOVebkEOHDjE7O8t1112Hz+fD5/Px1FNP8Y1vfAOfz0dfX9+2nTNp7KwzgUCA66+/nieeeMJ7zbZtnnjiCQ4cOLCBI5O0smvXLvr7+5vmKpfL8eyzz3pzdeDAAbLZLIcOHfI+8+STT2LbNjfffPMFH/N2RwjBpz/9aR555BGefPJJdu3a1fT+9ddfj9/vb5qz119/nfHx8aY5O3LkSJOR+vjjjxOPx9m/f/+F2ZGLGNu2qVQqcq42IbfccgtHjhzh8OHD3r8bbriBO++80/v/bTtnG62Q3o489NBDQtd18d3vflccO3ZMfOpTnxJdXV1N6nXJhSGfz4uXXnpJvPTSSwIQX/va18RLL70kxsbGhBBCPPDAA6Krq0v867/+q3j55ZfFBz/4QbFr1y5RLpe9bdx2223i2muvFc8++6z42c9+Jvbu3Ss+8pGPbNQubWv+4A/+QCQSCfGf//mfYmpqyvtXKpW8z/z+7/++GBkZEU8++aR44YUXxIEDB8SBAwe892u1mrjiiivE+9//fnH48GHx2GOPid7eXnHvvfduxC5ta77whS+Ip556Spw4cUK8/PLL4gtf+IJQFEX8x3/8hxBCztVWoDEbS4jtO2fS2DlPPPjgg2JkZEQEAgFx0003iZ///OcbPaSLkp/+9KcCWPLvrrvuEkI46edf/OIXRV9fn9B1Xdxyyy3i9ddfb9rG/Py8+MhHPiKi0aiIx+Pi4x//uMjn8xuwN9ufdnMFiH/6p3/yPlMul8Uf/uEfimQyKcLhsPjN3/xNMTU11bSdd955R9x+++0iFAqJnp4e8cd//MeiWq1e4L3Z/vzu7/6uGB0dFYFAQPT29opbbrnFM3SEkHO1FWg1drbrnClCCLExPiWJRCKRSCSS84/U7EgkEolEItnWSGNHIpFIJBLJtkYaOxKJRCKRSLY10tiRSCQSiUSyrZHGjkQikUgkkm2NNHYkEolEIpFsa6SxI5FIJBKJZFsjjR2JRCKRSCTbGmnsSCSSLcHHPvYxFEXx/nV3d3Pbbbfx8ssvb/TQJBLJJkcaOxKJZMtw2223MTU1xdTUFE888QQ+n49f//Vf3+hhSSSSTY40diQSyZZB13X6+/vp7+/nmmuu4Qtf+AITExPMzc0B8Kd/+qfs27ePcDjMJZdcwhe/+EWq1ar3/V/84he8733vIxaLEY/Huf7663nhhRcAGBsb4wMf+ADJZJJIJMLll1/Oj3/84w3ZT4lEsr74NnoAEolEcjYUCgW+973vsWfPHrq7uwGIxWJ897vfZXBwkCNHjvDJT36SWCzGn/zJnwBw5513cu211/Ktb30LTdM4fPgwfr8fgLvvvhvTNHn66aeJRCIcO3aMaDS6YfsnkUjWD9kIVCKRbAk+9rGP8b3vfY9gMAhAsVhkYGCARx99lOuuu67td/7qr/6Khx56yPPexONxHnzwQe66664ln73qqqv48Ic/zH333Xf+dkIikWwIMowlkUi2DO973/s4fPgwhw8f5rnnnuPWW2/l9ttvZ2xsDIDvf//7vPvd76a/v59oNMqf//mfMz4+7n3/nnvu4ROf+AQHDx7kgQce4K233vLe+6M/+iP+4i/+gne/+93cd999UvgskWwjpLEjkUi2DJFIhD179rBnzx5uvPFGvvOd71AsFvmHf/gHnnnmGe68805+7dd+jUcffZSXXnqJP/uzP8M0Te/7X/7ylzl69Ch33HEHTz75JPv37+eRRx4B4BOf+ARvv/02v/M7v8ORI0e44YYbePDBBzdqVyUSyToiw1gSiWRL8LGPfYxsNssPf/hD7zXbtunq6uKTn/wkg4OD/N3f/V2Tt+YTn/gEP/jBD8hms223+ZGPfIRisci//du/LXnv3nvv5Uc/+pH08Egk2wApUJZIJFuGSqXC9PQ0AAsLC/zt3/4thUKBD3zgA+RyOcbHx3nooYe48cYb+dGPfuR5bQDK5TKf//zn+a3f+i127drFyZMnef755/nwhz8MwGc/+1luv/129u3bx8LCAj/96U+57LLLNmQ/JRLJ+iKNHYlEsmV47LHHGBgYAJzMq0svvZSHH36Y9773vQB87nOf49Of/jSVSoU77riDL37xi3z5y18GQNM05ufn+ehHP8rMzAw9PT186EMf4itf+QoAlmVx9913c/LkSeLxOLfddht//dd/vRG7KZFI1hkZxpJIJBKJRLKtkQJliUQikUgk2xpp7EgkEolEItnWSGNHIpFIJBLJtkYaOxKJRCKRSLY10tiRSCQSiUSyrZHGjkQikUgkkm2NNHYkEolEIpFsa6SxI5FIJBKJZFsjjR2JRCKRSCTbGmnsSCQSiUQi2dZIY0cikUgkEsm2Rho7EolEIpFItjX/P4gBw0sVcDGkAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# make a scatter plot of the bass and high features\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.scatter(data['bass'], data['high'], alpha=0.2, s=2)\n",
    "\n",
    "# add a line at the mean of the bass feature\n",
    "#plt.axvline(data['bass'].mean(), color='red', linestyle='--', alpha=0.5)\n",
    "# add a line at one standard deviation away from the mean of the bass feature\n",
    "plt.axvline(data['bass'].mean() + 2 * data['bass'].std(), color='red', linestyle='--', alpha=1)\n",
    "plt.axvline(data['bass'].mean() - 2 * data['bass'].std(), color='red', linestyle='--', alpha=1)\n",
    "\n",
    "# add a line at the mean of the high feature\n",
    "#plt.axhline(data['high'].mean(), color='blue', linestyle='--', alpha=0.5)\n",
    "# add a line at one standard deviation away from the mean of the high feature\n",
    "plt.axhline(data['high'].mean() + 2 * data['high'].std(), color='blue', linestyle='--', alpha=1)\n",
    "plt.axhline(data['high'].mean() - 2 * data['high'].std(), color='blue', linestyle='--', alpha=1)\n",
    "\n",
    "plt.xlabel('Bass')\n",
    "plt.ylabel('High')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "bass above mean: 1531\n",
      "bass between mean: 47243\n",
      "bass below mean: 1226\n",
      "\n",
      "mid above mean: 1394\n",
      "mid between mean: 47503\n",
      "mid below mean: 1103\n",
      "\n",
      "high above mean: 1262\n",
      "high between mean: 47643\n",
      "high below mean: 1095\n",
      "\n"
     ]
    }
   ],
   "source": [
    "alpha = 2\n",
    "import numpy as np\n",
    "# now we need to classify each file based on the features\n",
    "# we will compare to the mean, and look its its above alpha stds away from the mean or below \n",
    "# -1 if below, 0 if between, 1 if above\n",
    "\n",
    "new_data = data.copy()\n",
    "\n",
    "for feature in ['bass', 'mid', 'high']:\n",
    "    new_data[f'{feature}_class'] = np.where(data[feature] > data[feature].mean() + alpha * data[feature].std(), 1, np.where(data[feature] < data[feature].mean() - alpha * data[feature].std(), -1, 0))\n",
    "\n",
    "new_data.head()\n",
    "new_data.describe()\n",
    "\n",
    "# count how many files are above and below the mean\n",
    "for feature in ['bass', 'mid', 'high']:\n",
    "    print(f\"{feature} above mean: {new_data[new_data[feature + '_class'] == 1].shape[0]}\")\n",
    "    print(f\"{feature} between mean: {new_data[new_data[feature + '_class'] == 0].shape[0]}\")\n",
    "    print(f\"{feature} below mean: {new_data[new_data[feature + '_class'] == -1].shape[0]}\")\n",
    "    print()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "bass above mean:\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/cbfd2e00-6f6d-43f8-8160-f3a3187d71b7.wav 417.1603088378906\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/e7aacce6-ebb6-4e50-a543-b190284d0869.wav 411.14306640625\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/b67c96b9-9ad2-46e0-8c29-5f52bd783338.wav 410.3070983886719\n",
      "bass below mean:\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/a3d70f80-98ff-4e40-a235-c08cf7b5e672.wav 0.4778965413570404\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/d59a4dc9-989a-484c-956f-52dccf057d69.wav 0.5737006068229675\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/b110e62e-318f-4507-86a9-c46e0bf213d4.wav 0.6637535095214844\n",
      "\n",
      "mid above mean:\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/2e383417-306e-43b3-8172-03eb912be270.wav 211.9332733154297\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/383ddb42-031d-48ed-89f9-0518d7e45c8f.wav 210.9254608154297\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/8542017c-28dd-4b0e-8eca-c41ecaabe282.wav 206.9360809326172\n",
      "mid below mean:\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/a3d70f80-98ff-4e40-a235-c08cf7b5e672.wav 1.2385283708572388\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/9038eb95-2a8a-42da-93ef-e25baaa2beec.wav 4.905820369720459\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/fd1666cc-c529-4123-9442-263b2d330bb0.wav 11.877724647521973\n",
      "\n",
      "high above mean:\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/d59a4dc9-989a-484c-956f-52dccf057d69.wav 133.80149841308594\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/bbe4d289-b61a-4bf1-9c78-4c7d5252a6cc.wav 120.52124786376953\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/9d566c32-53e1-4ef8-b172-35f23d53831e.wav 117.1395492553711\n",
      "high below mean:\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/d9dcd7cd-e867-4bc7-893e-2fd726555723.wav 0.06754320859909058\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/2ad6fc07-ff8e-45cd-bb0b-ba54bc8045d1.wav 0.07881976664066315\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/aeea865a-69c1-407a-8694-cc6f6632c850.wav 0.3136812746524811\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# lets list the first 3 files that are above and below the mean\n",
    "# find the filepaths with the highest and lowest values for each feature\n",
    "for feature in ['bass', 'mid', 'high']:\n",
    "    print(f\"{feature} above mean:\")\n",
    "    filepaths = new_data[new_data[feature + '_class'] == 1]['filepath'].tolist()\n",
    "    filepaths = sorted(filepaths, key=lambda x: new_data[new_data['filepath'] == x][feature].values[0], reverse=True)\n",
    "    for filepath in filepaths[:3]:\n",
    "        print(filepath, new_data[new_data['filepath'] == filepath][feature].values[0])\n",
    "    print(f\"{feature} below mean:\")\n",
    "    filepaths = new_data[new_data[feature + '_class'] == -1]['filepath'].tolist()\n",
    "    filepaths = sorted(filepaths, key=lambda x: new_data[new_data['filepath'] == x][feature].values[0])\n",
    "    for filepath in filepaths[:3]:\n",
    "        print(filepath, new_data[new_data['filepath'] == filepath][feature].values[0])\n",
    "    print()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/fbcd1c15-db1f-4b01-8965-39db7299bff3.wav 399.7385559082031 2.0149240493774414\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/5c4adb80-7d7e-4907-aa26-752f9339fbf0.wav 313.93023681640625 2.0499162673950195\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/2f481cd2-2374-472e-b7a7-808644eea830.wav 285.56329345703125 4.800220012664795\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/19bad336-a4a7-4813-907c-d30a5b8715f0.wav 284.16656494140625 5.317797660827637\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/8bdf7a8e-9502-4219-8cec-3713e5ac64ae.wav 288.30657958984375 5.4228835105896\n"
     ]
    }
   ],
   "source": [
    "# can we also look for files that are above in bass and below in high?\n",
    "filepaths = new_data[new_data['bass_class'] == 1]['filepath'].tolist()\n",
    "filepaths = sorted(filepaths, key=lambda x: new_data[new_data['filepath'] == x]['high'].values[0])\n",
    "for filepath in filepaths[:5]:\n",
    "    print(filepath, new_data[new_data['filepath'] == filepath]['bass'].values[0], new_data[new_data['filepath'] == filepath]['high'].values[0])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2757\n",
      "2497\n",
      "2357\n"
     ]
    }
   ],
   "source": [
    "# how many files have a non-zero value for any of the features?\n",
    "print(new_data[new_data['bass_class'] != 0].shape[0])\n",
    "print(new_data[new_data['mid_class'] != 0].shape[0])\n",
    "print(new_data[new_data['high_class'] != 0].shape[0])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [],
   "source": [
    "def classify_spectrum(bass_class, mid_class, high_class):\n",
    "    # First, let's handle the perfectly balanced case\n",
    "    if (bass_class, mid_class, high_class) == (0, 0, 0):\n",
    "        return \"Balanced\"\n",
    "    \n",
    "    # Handle common patterns based on the most dominant characteristics\n",
    "    # Bass characteristics\n",
    "    if bass_class == 1:\n",
    "        if high_class == -1:\n",
    "            return \"Dark/Warm\"\n",
    "        if mid_class <= 0:\n",
    "            return \"Boomy\"\n",
    "        return \"Bass-heavy\"\n",
    "            \n",
    "    if bass_class == -1:\n",
    "        if high_class >= 0:\n",
    "            return \"Tinny\"\n",
    "        return \"Thin\"\n",
    "    \n",
    "    # Mid-focused characteristics\n",
    "    if mid_class == 1:\n",
    "        if high_class <= 0:\n",
    "            return \"Boxy\"\n",
    "        return \"Forward\"\n",
    "    \n",
    "    if mid_class == -1:\n",
    "        if high_class == 1:\n",
    "            return \"Scooped\"\n",
    "        return \"Hollow\"\n",
    "    \n",
    "    # High-end characteristics\n",
    "    if high_class == 1:\n",
    "        return \"Bright\"\n",
    "    if high_class == -1:\n",
    "        return \"Dull\"\n",
    "    \n",
    "    # If we get here, it's a subtle case\n",
    "    return \"Neutral\"\n",
    "# Example usage:\n",
    "# print(classify_spectrum(-1, -1, 1))  # Would return \"Tinny\"\n",
    "# print(classify_spectrum(1, 0, 0))    # Would return \"Warm\"\n",
    "\n",
    "new_data['spectrum'] = new_data.apply(lambda row: classify_spectrum(row['bass_class'], row['mid_class'], row['high_class']), axis=1)\n",
    "\n",
    "# save csv of the new_data\n",
    "new_data.to_csv('audio_production_features_v2.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# create a bar plot for each of the spectrum\n",
    "new_data['spectrum'].value_counts().plot(kind='bar')\n",
    "# put the values on top of the bars\n",
    "for i, v in enumerate(new_data['spectrum'].value_counts()):\n",
    "    plt.text(i, v, str(v), ha='center', va='bottom')\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/e9c185f0-b102-406f-83c8-eb6e6d269135.wav\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/188bf484-dfda-4d00-bf28-b271f21ddf17.wav\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/c9d657df-6a6a-4278-b350-3bb2718e32a6.wav\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/5ba8139d-881d-4e73-8cfb-1988c81df598.wav\n",
      "/app/suno/data/audio_2ch_48khz_lg/train/genius_hq/0e588813-31be-42b8-a4f8-6ba4f47fc9a4.wav\n"
     ]
    }
   ],
   "source": [
    "# get a list of all the thin fileptahs\n",
    "thin_filepaths = new_data[new_data['spectrum'] == 'Dark/Warm']['filepath'].tolist()\n",
    "for filepath in thin_filepaths[:5]:\n",
    "    print(filepath)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "500\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "bass 161.9 (43.0) mid 118.0 (22.3) high 50.4 (12.9):   9%|▉         | 47/500 [00:36<05:55,  1.27it/s]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[32], line 34\u001b[0m\n\u001b[1;32m     31\u001b[0m     x \u001b[38;5;241m=\u001b[39m torchaudio\u001b[38;5;241m.\u001b[39mfunctional\u001b[38;5;241m.\u001b[39mresample(x, sr, SAMPLE_RATE)\n\u001b[1;32m     33\u001b[0m \u001b[38;5;66;03m# loudness normalization\u001b[39;00m\n\u001b[0;32m---> 34\u001b[0m loudness \u001b[38;5;241m=\u001b[39m \u001b[43mmeter\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mintegrated_loudness\u001b[49m\u001b[43m(\u001b[49m\u001b[43mx\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpermute\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mnumpy\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     35\u001b[0m x \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m10\u001b[39m \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39m ((TARGET_LOUDNESS \u001b[38;5;241m-\u001b[39m loudness) \u001b[38;5;241m/\u001b[39m \u001b[38;5;241m20.0\u001b[39m)\n\u001b[1;32m     37\u001b[0m \u001b[38;5;66;03m# compute features\u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda3/envs/suno_env/lib/python3.10/site-packages/pyloudnorm/meter.py:66\u001b[0m, in \u001b[0;36mMeter.integrated_loudness\u001b[0;34m(self, data)\u001b[0m\n\u001b[1;32m     64\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m (filter_class, filter_stage) \u001b[38;5;129;01min\u001b[39;00m iteritems(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_filters):\n\u001b[1;32m     65\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m ch \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(numChannels):\n\u001b[0;32m---> 66\u001b[0m         input_data[:,ch] \u001b[38;5;241m=\u001b[39m \u001b[43mfilter_stage\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mapply_filter\u001b[49m\u001b[43m(\u001b[49m\u001b[43minput_data\u001b[49m\u001b[43m[\u001b[49m\u001b[43m:\u001b[49m\u001b[43m,\u001b[49m\u001b[43mch\u001b[49m\u001b[43m]\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     68\u001b[0m G \u001b[38;5;241m=\u001b[39m [\u001b[38;5;241m1.0\u001b[39m, \u001b[38;5;241m1.0\u001b[39m, \u001b[38;5;241m1.0\u001b[39m, \u001b[38;5;241m1.41\u001b[39m, \u001b[38;5;241m1.41\u001b[39m] \u001b[38;5;66;03m# channel gains\u001b[39;00m\n\u001b[1;32m     69\u001b[0m T_g \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mblock_size \u001b[38;5;66;03m# 400 ms gating block standard\u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda3/envs/suno_env/lib/python3.10/site-packages/pyloudnorm/iirfilter.py:170\u001b[0m, in \u001b[0;36mIIRfilter.apply_filter\u001b[0;34m(self, data)\u001b[0m\n\u001b[1;32m    157\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mapply_filter\u001b[39m(\u001b[38;5;28mself\u001b[39m, data):\n\u001b[1;32m    158\u001b[0m \u001b[38;5;250m    \u001b[39m\u001b[38;5;124;03m\"\"\" Apply the IIR filter to an input signal.\u001b[39;00m\n\u001b[1;32m    159\u001b[0m \n\u001b[1;32m    160\u001b[0m \u001b[38;5;124;03m    Params\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    168\u001b[0m \u001b[38;5;124;03m        Filtered input audio.\u001b[39;00m\n\u001b[1;32m    169\u001b[0m \u001b[38;5;124;03m    \"\"\"\u001b[39;00m\n\u001b[0;32m--> 170\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mpassband_gain \u001b[38;5;241m*\u001b[39m \u001b[43mscipy\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msignal\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mlfilter\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mb\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43ma\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdata\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m~/miniconda3/envs/suno_env/lib/python3.10/site-packages/scipy/signal/_signaltools.py:2156\u001b[0m, in \u001b[0;36mlfilter\u001b[0;34m(b, a, x, axis, zi)\u001b[0m\n\u001b[1;32m   2154\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m   2155\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m zi \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m-> 2156\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43m_sigtools\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_linear_filter\u001b[49m\u001b[43m(\u001b[49m\u001b[43mb\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43ma\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mx\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   2157\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m   2158\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m _sigtools\u001b[38;5;241m.\u001b[39m_linear_filter(b, a, x, axis, zi)\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "import glob\n",
    "import os\n",
    "from tqdm import tqdm\n",
    "import pyloudnorm as pyln\n",
    "import numpy as np\n",
    "SAMPLE_RATE = 48000\n",
    "TARGET_LOUDNESS = -16.0\n",
    "\n",
    "#filepaths = glob.glob(\"/home/christian/audio/reference-audio-wav/*.wav\")\n",
    "filepaths = glob.glob(\"/app/suno/data/audio_2ch_48khz_lg/val/genius_hq/*.wav\")\n",
    "print(len(filepaths))\n",
    "meter = pyln.Meter(SAMPLE_RATE)\n",
    "\n",
    "filepaths = filepaths\n",
    "\n",
    "features = {\n",
    "    'bass': [],\n",
    "    'mid': [],\n",
    "    'high': []\n",
    "}\n",
    "\n",
    "pbar = tqdm(filepaths)\n",
    "for filepath in pbar:\n",
    "    x, sr = torchaudio.load(filepath)\n",
    "\n",
    "    # for speed, only use first 60s\n",
    "    x = x[:, :int(60 * sr)]\n",
    "\n",
    "    # resample if necessary\n",
    "    if sr != SAMPLE_RATE:\n",
    "        x = torchaudio.functional.resample(x, sr, SAMPLE_RATE)\n",
    "\n",
    "    # loudness normalization\n",
    "    loudness = meter.integrated_loudness(x.permute(1, 0).numpy())\n",
    "    x *= 10 ** ((TARGET_LOUDNESS - loudness) / 20.0)\n",
    "    \n",
    "    # compute features\n",
    "    results = compute_all_bands(x, SAMPLE_RATE)\n",
    "\n",
    "    for key, value in results.items():\n",
    "        features[key].append(value)\n",
    "\n",
    "    pbar.set_description(f\"bass {np.mean(features['bass']):.1f} ({np.std(features['bass']):.1f}) mid {np.mean(features['mid']):.1f} ({np.std(features['mid']):.1f}) high {np.mean(features['high']):.1f} ({np.std(features['high']):.1f})\")\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# make a scatter plot of the bass and high features\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.scatter(features['bass'], features['high'])\n",
    "plt.xlabel('Bass')\n",
    "plt.ylabel('High')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "suno_env",
   "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": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}