{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Guide Calling Using the Cellranger Module HDF5 File \n",
    "## Utilising Direct Capture Perturb-Seq to Deciphering Alternative Promoter Usage  \n",
    "\n",
    "1. Run through the 10x Cellranger pipeline and velocyto for single cell RNAseq quatification and using (2) guides quantifiction. all found in the cellranger files folder **bash**\n",
    "2.  Guide Calling for dual guide. Use repogle method to take molecule.h5 generated by cellranger and py to run through repogle version of guide calling or use cellranger_guidecalling.ipynb for Direct Capture Perturb-Seq dual guide. Formed guide-specific lists of cells.\n",
    "3. Pseudobulk analysis.\n",
    "    A. Seperation of guide-specific fastq files. **bash**\n",
    "    B. Whippet pseudobulk for transcript specific analysis, post UMI deduplication.  **bash**\n",
    "    C. Transcript quality control. **R**\n",
    "    D. Whippet result visualisation.\n",
    "4. Normalisation of adata object and E-distance of KD\n",
    "5. Check gene and neighboring gene expression\n",
    "6. Create individual umaps per gene of interest\n",
    "    A. UMAPs \n",
    "    B. Rand Index score\n",
    "7. Cell phase assignment model from FUCCI-matched single cell paper (GSE146773)\n",
    "8. Differential Expression analysis.\n",
    "    A. Find the shared P1 and P2 genes. \n",
    "    B. Check the shared P1 and P2 across different protospacers with the same A/B and C/D.\n",
    "9.  **CNV Score & Numbat to quantify and Velocity quantification with loom file**\n",
    "10. ESR1-specific analysis from proliferation analysis to rt-qpcr\n",
    "11.  Spectra analysis and visualisation for pathway enrichment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/helenking/anaconda3/envs/apu/lib/python3.12/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Scanpy 1.10.3\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%matplotlib inline\n",
    "%autoreload 2\n",
    "#general\n",
    "import scanpy as sc\n",
    "import matplotlib.pyplot as pl\n",
    "import anndata as ad\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import hdf5plugin\n",
    "\n",
    "#form a location\n",
    "loc=\"../../../alt-prom-crispr-fiveprime/\"\n",
    "\n",
    "import seaborn as sns\n",
    "from tqdm.notebook import tqdm\n",
    "import scperturb\n",
    "import sys\n",
    "sys.path.append(loc+'scripts/')\n",
    "from apu_analysis import *\n",
    "import infercnvpy as cnv\n",
    "from apu_analysis.cell_import import CellPopulation\n",
    "from IPython.display import clear_output\n",
    "pd.options.display.float_format = '{:.4f}'.format\n",
    "import matplotlib.pyplot as plt\n",
    "import scvelo as scv\n",
    "from matplotlib_venn import venn3\n",
    "\n",
    "#for this python\n",
    "from scipy.special import rel_entr\n",
    "import sklearn.cluster as cluster\n",
    "import umap\n",
    "from scipy import stats\n",
    "from scipy.stats import bootstrap\n",
    "import statsmodels.api as sm\n",
    "from statsmodels.stats.multitest import multipletests\n",
    "from numpy import reshape\n",
    "from numpy import array\n",
    "from sklearn.decomposition import PCA\n",
    "import os \n",
    "from sklearn.ensemble import IsolationForest\n",
    "from sklearn.covariance import EllipticEnvelope\n",
    "from sklearn.neighbors import LocalOutlierFactor\n",
    "from sklearn.svm import OneClassSVM\n",
    "import seaborn as sns\n",
    "import statsmodels.api as sm\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "# colours using garvan \n",
    "color1 ='#4d00c7'\n",
    "palecolor1=\"#b366ff\"\n",
    "color2= '#da3c07'\n",
    "palecolor2=\"#ff8954\"\n",
    "color3='#05d3d3'\n",
    "color4='#c6c7c5'\n",
    "color4=\"#434541\"\n",
    "color5=\"#eb31e1\"\n",
    "color6=\"#3175eb\"\n",
    "color7=\"#a7eb31\"\n",
    "color8=\"#b366ff\"\n",
    "color9=\"#ff8954\"\n",
    "color10=\"#35c9d4\"\n",
    "\n",
    "#use viridis\n",
    "color1=\"#fde725\"\n",
    "color2=\"#7ad151\"\n",
    "color3=\"#22a884\"\n",
    "color4=\"#2a788e\"\n",
    "color5=\"#2a788e\"\n",
    "color6='#440154'\n",
    "\n",
    "\n",
    "# Create the color palette\n",
    "palette = sns.color_palette([palecolor1,palecolor2])\n",
    "palette2 = sns.color_palette([color1, color2, color3, color4,color5,color6 ,color7])\n",
    "palette = sns.color_palette([color1, color2,color3])\n",
    "new_palette = sns.color_palette([color1, color2,color1, color2,color1, color2,color1, color2,color1, color2,color1, color2, color3, color4])\n",
    "\n",
    "\n",
    "\n",
    "print(\"Scanpy\", sc.__version__)\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "adata = ad.read_h5ad(loc+\"files/adata_normalised_cellcycle.h5ad\")\n",
    "adata.X = adata.layers[\"counts\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.\n"
     ]
    }
   ],
   "source": [
    "#calculate leiden\n",
    "sc.tl.pca(adata, n_comps=50, svd_solver='arpack')\n",
    "sc.pp.neighbors(adata, n_neighbors=10, n_pcs=50)\n",
    "sc.tl.leiden(adata, resolution=0.5, key_added=\"leiden\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import statsmodels.api as sm\n",
    "import statsmodels.formula.api as smf\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "def run_single_guide_regression(adataControl, controlGuides, guideIndex, expressionMatrix, method='NB'):\n",
    "    \"\"\"\n",
    "    Test whether a single control guide induces transcriptional changes across genes.\n",
    "    \n",
    "    Parameters:\n",
    "    - adataControl: AnnData object filtered to only include control guide cells.\n",
    "    - controlGuides: List of control guide names.\n",
    "    - guideIndex: Index of the control guide to test.\n",
    "    - expressionMatrix: DataFrame of gene expression (cells x genes).\n",
    "    - method: 'NB' or 'OLS'\n",
    "    \n",
    "    Returns:\n",
    "    - DataFrame with results per gene (Gene, Coef, StdErr, Pval)\n",
    "    \"\"\"\n",
    "    guide_name = controlGuides[guideIndex]\n",
    "    print(f\"Running regression for guide: {guide_name}\")\n",
    "    \n",
    "    # Create binary predictor: 1 = cell has this guide, 0 = other control guides\n",
    "    covariate_df = adataControl.obs[[\"guide_identity\", \"n_genes_by_counts\", \"pct_counts_mt\", \"leiden\"]].copy()\n",
    "    covariate_df[\"is_guide\"] = (covariate_df[\"guide_identity\"] == guide_name).astype(int)\n",
    "    \n",
    "    results = []\n",
    "    for gene in expressionMatrix.columns:\n",
    "        covariate_df[\"y\"] = expressionMatrix[gene].values\n",
    "        \n",
    "        formula = \"y ~ is_guide + n_genes_by_counts + pct_counts_mt + C(leiden)\"  # Use C() for categorical leiden\n",
    "        try:\n",
    "            if method == 'NB':\n",
    "                model = smf.glm(formula=formula, data=covariate_df,\n",
    "                                family=sm.families.NegativeBinomial()).fit()\n",
    "            elif method == 'OLS':\n",
    "                model = smf.ols(formula=formula, data=covariate_df).fit()\n",
    "            else:\n",
    "                raise ValueError(\"Method must be 'NB' or 'OLS'\")\n",
    "            \n",
    "            coef = model.params.get(\"is_guide\", np.nan)\n",
    "            pval = model.pvalues.get(\"is_guide\", np.nan)\n",
    "            stderr = model.bse.get(\"is_guide\", np.nan)\n",
    "            results.append((gene, coef, stderr, pval))\n",
    "        except Exception as e:\n",
    "            results.append((gene, np.nan, np.nan, np.nan))\n",
    "\n",
    "    res_df = pd.DataFrame(results, columns=[\"Gene\", \"Coef\", \"StdErr\", \"Pval\"])\n",
    "    res_df[\"Guide\"] = guide_name\n",
    "    return res_df\n",
    "\n",
    "\n",
    "\n",
    "    \"\"\"\n",
    "    Create a volcano plot for regression results.\n",
    "    \"\"\"\n",
    "    res_df = res_df.copy()\n",
    "    res_df[\"-log10Pval\"] = -np.log10(res_df[\"Pval\"])\n",
    "    \n",
    "    plt.figure(figsize=(8, 6))\n",
    "    sns.scatterplot(data=res_df, x=\"Coef\", y=\"-log10Pval\", alpha=0.6, edgecolor=None)\n",
    "\n",
    "    # Highlight significant hits\n",
    "    sig_hits = res_df[(res_df[\"Pval\"] < pval_threshold) & (res_df[\"Coef\"].abs() > lfc_threshold)]\n",
    "    plt.scatter(sig_hits[\"Coef\"], sig_hits[\"-log10Pval\"], color=\"red\", alpha=0.8)\n",
    "\n",
    "    plt.axhline(-np.log10(pval_threshold), color='gray', linestyle='--', label=f'p={pval_threshold}')\n",
    "    plt.axvline(-lfc_threshold, color='gray', linestyle='--')\n",
    "    plt.axvline(lfc_threshold, color='gray', linestyle='--')\n",
    "\n",
    "    plt.title(f\"Volcano Plot for Guide: {guide_name}\")\n",
    "    plt.xlabel(\"Coefficient (effect size)\")\n",
    "    plt.ylabel(\"-log10(p-value)\")\n",
    "    plt.grid(True)\n",
    "    plt.tight_layout()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "# STEP 1: Define control guide cells (e.g. NTC or CTRL in identity)\n",
    "#get all guides names into a list for adata.obs['guide_target'][].unique(\n",
    "controlGuides = adata.obs['guide_identity'][adata.obs['guide_identity'].str.startswith('non-tar')].unique().tolist()\n",
    "control_mask = adata.obs['guide_identity'].isin(controlGuides[5:25])  # Adjust this to select your control guides\n",
    "adataControl = adata[control_mask].copy()\n",
    "#filter for first twenty guides \n",
    "# STEP 2: Prepare expression matrix\n",
    "\n",
    "expressionMatrix = pd.DataFrame(adataControl.layers[\"log1p\"])\n",
    "expressionMatrix.columns = adata.var_names\n",
    "expressionMatrix.index = adataControl.obs_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pca = PCA(n_components=100)\n",
    "pcs = pca.fit_transform(expressionMatrix)  # shape = (n_cells, 100)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "encoder = OneHotEncoder(sparse_output=False)\n",
    "design_matrix = encoder.fit_transform(adataControl.obs[['guide_assignment']])\n",
    "guide_names = encoder.categories_[0]  # guide column names\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "betas = []\n",
    "for i in range(pcs.shape[1]):  # for each PC\n",
    "    model = sm.OLS(pcs[:, i], sm.add_constant(design_matrix)).fit()\n",
    "    betas.append(model.params[1:])  # skip intercept\n",
    "\n",
    "coeff_matrix = np.vstack(betas).T  # shape = (n_guides x n_PCs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "scaled = StandardScaler().fit_transform(coeff_matrix)\n",
    "\n",
    "# Select top 20 PCs (columns)\n",
    "sns.clustermap(pd.DataFrame(scaled[:, :20], index=guide_names),\n",
    "               cmap=\"vlag\", metric=\"euclidean\", figsize=(12, 10),\n",
    "               xticklabels=[f\"PC{i+1}\" for i in range(20)],\n",
    "                yticklabels=[f\"non-targeting control {i+1}\" for i in range(20)])\n",
    "\n",
    "plt.title(\"Guide × PC Coefficient Heatmap (top 20 PCs)\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running model: IsolationForest\n",
      "Running model: EllipticEnvelope\n",
      "Running model: LocalOutlierFactor\n",
      "Running model: OneClassSVM\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "models = {\n",
    "    \"IsolationForest\": IsolationForest(random_state=0, contamination=0.1),\n",
    "    \"EllipticEnvelope\": EllipticEnvelope(contamination=0.1),\n",
    "    \"LocalOutlierFactor\": LocalOutlierFactor(),\n",
    "    \"OneClassSVM\": OneClassSVM( nu=0.01)\n",
    "}\n",
    "c=0\n",
    "outlier_scores = {}\n",
    "#subset coeff_matrix array to be only first 20 \n",
    "for name, model in models.items():\n",
    "    print(f\"Running model: {name}\")\n",
    "    \n",
    "    if name == \"LocalOutlierFactor\":\n",
    "        pred = model.fit_predict(coeff_matrix)\n",
    "    else:\n",
    "        model.fit(coeff_matrix)\n",
    "        pred = model.predict(coeff_matrix)\n",
    "    #add the votes with the name to the guide\n",
    "    #if reaching 20 then break\n",
    "    outlier_scores[name] = pd.Series(pred, index=guide_names, name=name)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "outlier_counts_side = pd.DataFrame(outlier_scores).apply(lambda x: (x == -1).sum(), axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "IsolationForest        0\n",
       "EllipticEnvelope       2\n",
       "LocalOutlierFactor     0\n",
       "OneClassSVM           20\n",
       "dtype: int64"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "outlier_counts_side"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA9sAAAJOCAYAAACnVRSYAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAACMFElEQVR4nOzdeZyN9f//8eeZwdiyL4OUnexZsxWiQYW0SIslaeMTzSeiROrzSfUpFApRtKJCO0qUylZItr7WrEP2DGbMzOv3R785zWEwo7O9Zx732+3cONd1nes8z3ve5z3zOue63pfHzEwAAAAAAMBvIkIdAAAAAACArIZiGwAAAAAAP6PYBgAAAADAzyi2AQAAAADwM4ptAAAAAAD8jGIbAAAAAAA/o9gGAAAAAMDPKLYBAAAAAPAzim0AAAAAAPyMYhsA4LSpU6fK4/Fo+/bt3mUtW7ZUy5YtQ5Ypu2jZsqVq1qwZ6hj/SM+ePVWuXLkLbrd9+3Z5PB5NnTo14JkAAFkDxTYAwK/WrVunu+66S2XKlFFUVJRKly6tO++8U+vWrftH+3322Wc1Z84c/4QMgNmzZ6t9+/YqVqyYcuXKpdKlS+u2227TN998E+pokqQ9e/boqaee0urVq0Md5aIkJCRo7Nixat68uQoXLuxt444dO+r9999XcnJyqCMCAOAjR6gDAACyjlmzZqlbt24qUqSIevfurfLly2v79u2aMmWKPvzwQ02fPl033XTTRe372Wef1S233KLOnTtfcNv58+df1HNcDDPTPffco6lTp+rKK69UbGysoqOjtXfvXs2ePVvXXnutfvjhBzVt2jRomdKzZ88ejRgxQuXKlVPdunVDmiWz/vjjD7Vv314///yzYmJiNHToUBUpUkRxcXH6+uuvdccdd2jz5s168sknM73v119/XSkpKQFIDQDI7ii2AQB+sWXLFt19992qUKGCvvvuOxUvXty7rn///mrRooXuvvturVmzRhUqVAholly5cvltXykpKUpMTFTu3LnTXf/SSy9p6tSpGjBggEaNGiWPx+Nd98QTT+jtt99Wjhz8uv0n7r77bq1atUofffSRunTp4rNuyJAh+umnn/Tbb79d1L5z5szpj4gAAJyFw8gBAH7xv//9TydOnNCkSZN8Cm1JKlasmCZOnKj4+Hi98MIL3uXnOl/2qaee8ilaPR6P4uPjNW3aNHk8Hnk8HvXs2fOcWdI7ZzshIUHDhw9XpUqVFBUVpbJly2rQoEFKSEjw2c7j8ahfv3569913VaNGDUVFRWnu3LnpPs/Jkyc1cuRIVatWTS+++KJP5lR33323GjVq5L2/detW3XrrrSpSpIjy5s2rq666Sp9//rnPY9I7D12SFi1aJI/Ho0WLFvm81po1a2r9+vVq1aqV8ubNqzJlyvi086JFi9SwYUNJUq9evbxtmHr+8aZNm3TzzTcrOjpauXPn1qWXXqrbb79dR48eTfd1n+nnn39W06ZNlSdPHpUvX14TJkzwrjt+/Ljy5cun/v37n/W4Xbt2KTIyUiNHjjznvpcsWaJ58+bpvvvuO6vQTtWgQQPdeeed3vuZab/0+uCRI0fUs2dPFSxYUIUKFVKPHj105MiRdJ9748aNuuWWW1SkSBHlzp1bDRo00CeffHLO1wMAyD74qB0A4BeffvqpypUrpxYtWqS7/uqrr1a5cuXOKiwz4u2339a9996rRo0a6b777pMkVaxYMcOPT0lJUceOHfX999/rvvvu0xVXXKFff/1Vo0eP1v/93/+ddS74N998o5kzZ6pfv34qVqzYOSfQ+v7773Xo0CENGDBAkZGRF8yxb98+NW3aVCdOnNDDDz+sokWLatq0aerYsaM+/PDDiz7E/vDhw2rXrp26dOmi2267TR9++KEee+wx1apVS+3bt9cVV1yhp59+WsOGDdN9993n/Rk1bdpUiYmJiomJUUJCgv71r38pOjpau3fv1meffaYjR46oYMGCF3zuDh066LbbblO3bt00c+ZMPfjgg8qVK5fuuece5c+fXzfddJNmzJihUaNG+bTT+++/LzPzKZTP9Omnn0qS7rrrrotqm8wyM3Xq1Enff/+9HnjgAV1xxRWaPXu2evTocda269atU7NmzVSmTBkNHjxY+fLl08yZM9W5c2d99NFHF/3zBABkEQYAwD905MgRk2SdOnU673YdO3Y0SXbs2DEzM+vRo4ddfvnlZ203fPhwO/NXVL58+axHjx5nbfvmm2+aJNu2bZt32TXXXGPXXHON9/7bb79tERERtnjxYp/HTpgwwSTZDz/84F0mySIiImzdunXnfS1mZi+//LJJstmzZ19wWzOzAQMGmCSfHH/++aeVL1/eypUrZ8nJyed8TWZmCxcuNEm2cOFCn9cqyd566y3vsoSEBIuOjrabb77Zu2zFihUmyd58802ffa5atcok2QcffJCh15BW6nO/9NJLPs9dt25dK1GihCUmJpqZ2bx580ySffnllz6Pr127ts/PKT033XSTSbIjR474LD958qT98ccf3tvhw4e96zLTfmf2wTlz5pgke+GFF7zLkpKSrEWLFme137XXXmu1atWyU6dOeZelpKRY06ZNrXLlyud9XQCArI/DyAEA/9iff/4pSbrkkkvOu13q+mPHjgU8U1offPCBrrjiClWrVk0HDhzw3lq3bi1JWrhwoc/211xzjapXr37B/aa+jgu97lRffPGFGjVqpObNm3uX5c+fX/fdd5+2b9+u9evXZ/Ql+cifP7/PN7+5cuVSo0aNtHXr1gs+NvWb63nz5unEiROZfu4cOXLo/vvv93nu+++/X/v379fPP/8sSWrTpo1Kly6td99917vd2rVrtWbNmgt+Y53axvnz5/dZPmHCBBUvXtx7S9um/8QXX3yhHDly6MEHH/Qui4yM1L/+9S+f7Q4dOqRvvvlGt912m/78809vnzp48KBiYmK0adMm7d692y+ZAABuotgGAPxjqcVmatF9Lhktyv1t06ZNWrdunU9xVrx4cVWpUkWStH//fp/ty5cvn6H9FihQQNKFX3eq33//XVWrVj1r+RVXXOFdfzEuvfTSs84XL1y4sA4fPnzBx5YvX16xsbGaPHmyihUrppiYGI0fPz7D52uXLl1a+fLl81mW2q6p50xHRETozjvv1Jw5c7wF/bvvvqvcuXPr1ltvPe/+U/vK8ePHfZbffPPN+uqrr/TVV1+pdu3aGcqaEb///rtKlSp1VnF/5s9t8+bNMjM9+eSTZ/Wr4cOHSzq7XwEAshfO2QYA/GMFCxZUqVKltGbNmvNut2bNGpUpU8ZbpKY3oZgkv18zOSUlRbVq1dKoUaPSXV+2bFmf+3ny5MnQfqtVqyZJ+vXXXzN0SbKMymy7nOt8cTPL0PO99NJL6tmzpz7++GPNnz9fDz/8sEaOHKmlS5fq0ksvzVjoC+jevbv+97//ac6cOerWrZvee+893XDDDRc8Jzy1jdeuXatmzZp5l5ctW9b7cytcuLAOHDjgXReMfpV6ubBHH31UMTEx6W5TqVIlvz0fAMA9FNsAAL+44YYb9Prrr+v7779P95DexYsXa/v27T6HHBcuXDjdWZ7T+4b3XAVURlSsWFG//PKLrr322n+0nzM1b95chQsX1vvvv6/HH3/8gpOkXX755eleomrjxo3e9dJf7SLprLa52G++pQu3X61atVSrVi0NHTpUP/74o5o1a6YJEyboP//5z3kft2fPHsXHx/t8u/1///d/kuQzsVzNmjV15ZVX6t1339Wll16qHTt2aOzYsRfMfcMNN+i5557Tu+++61Nsn88/ab/LL79cCxYs0PHjx32+3T7z55Z6+bqcOXOqTZs2GcoFAMheOIwcAOAXAwcOVJ48eXT//ffr4MGDPusOHTqkBx54QHnz5tXAgQO9yytWrKijR4/6fCO+d+9ezZ49+6z958uX75yXX7qQ2267Tbt379brr79+1rqTJ08qPj7+ovabN29ePfbYY9qwYYMee+yxdL9Jfuedd7R8+XJJUocOHbR8+XItWbLEuz4+Pl6TJk1SuXLlvOeJp860/t1333m3S05O1qRJky4qpyRvMXxmGx47dkxJSUk+y2rVqqWIiIizLouWnqSkJE2cONF7PzExURMnTlTx4sVVv359n23vvvtuzZ8/X2PGjFHRokXVvn37C+6/WbNmatu2rSZNmqSPP/443W3ObPd/0n4dOnRQUlKSXnvtNZ/HnvnBQIkSJdSyZUtNnDhRe/fuPWs/f/zxxwWfCwCQtfHNNgDALypXrqxp06bpzjvvVK1atdS7d2+VL19e27dv15QpU3TgwAG9//77Ppfsuv322/XYY4/ppptu0sMPP6wTJ07otddeU5UqVbRy5Uqf/devX19ff/21Ro0apdKlS6t8+fJq3LhxhrLdfffdmjlzph544AEtXLhQzZo1U3JysjZu3KiZM2dq3rx5atCgwUW97oEDB2rdunV66aWXtHDhQt1yyy2Kjo5WXFyc5syZo+XLl+vHH3+UJA0ePFjvv/++2rdvr4cfflhFihTRtGnTtG3bNn300UeKiPjrM/AaNWroqquu0pAhQ3To0CEVKVJE06dPP6sozoyKFSuqUKFCmjBhgi655BLly5dPjRs31i+//KJ+/frp1ltvVZUqVZSUlKS3335bkZGRuvnmmy+439KlS+v555/X9u3bVaVKFc2YMUOrV6/WpEmTlDNnTp9t77jjDg0aNEizZ8/Wgw8+eNb6c3nnnXfUrl07de7cWe3bt1ebNm1UuHBhxcXF6euvv9Z3333nU7j/k/a78cYb1axZMw0ePFjbt29X9erVNWvWrHTPYR8/fryaN2+uWrVqqU+fPqpQoYL27dunJUuWaNeuXfrll18y9PoAAFlUaCdDBwBkNWvWrLFu3bpZqVKlLGfOnBYdHW3dunWzX3/9Nd3t58+fbzVr1rRcuXJZ1apV7Z133kn30l8bN260q6++2vLkyWOSvJcBy8ilv8zMEhMT7fnnn7caNWpYVFSUFS5c2OrXr28jRoywo0ePereTZH379s306/7www/tuuuusyJFiliOHDmsVKlS1rVrV1u0aJHPdlu2bLFbbrnFChUqZLlz57ZGjRrZZ599dtb+tmzZYm3atLGoqCgrWbKkPf744/bVV1+le+mvGjVqnPX49C6r9vHHH1v16tUtR44c3stYbd261e655x6rWLGi5c6d24oUKWKtWrWyr7/++oKvOfW5f/rpJ2vSpInlzp3bLr/8chs3btw5H9OhQweTZD/++OMF95/WyZMnbcyYMdakSRMrUKCA5ciRw6Kjo+2GG26wd99915KSkny2z2j7pddOBw8etLvvvtsKFChgBQsWtLvvvtt7ibQzL522ZcsW6969u0VHR1vOnDmtTJkydsMNN9iHH36YqdcHAMh6PGYZnD0FAADgH7rpppv066+/avPmzaGOAgBAQHHONgAACIq9e/fq888/19133x3qKAAABBznbAMAgIDatm2bfvjhB02ePFk5c+b0mZEeAICsim+2AQBAQH377be6++67tW3bNk2bNk3R0dGhjgQAQMBxzjYAAAAAAH7GN9sAAAAAAPgZxTYAAAAAAH7GBGnpSElJ0Z49e3TJJZfI4/GEOg4AAAAAZGlmpj///FOlS5dWRETW+E6YYjsde/bsUdmyZUMdAwAAAACylZ07d+rSSy8NdQy/oNhOxyWXXCLprx90gQIFQpwGAAAAALK2Y8eOqWzZst5aLCug2E5H6qHjBQoUoNgGAAAAgCDJSqfxZo2D4QEAAAAACCMU2wAAAAAA+BnFNgAAAAAAfkaxDQAAAACAn1FsAwAAAADgZxTbAAAAAAD4GcU2AAAAAAB+RrENAAAAAICfUWwDAAAAAOBnFNsAAAAAAPgZxTYAAAAAAH4W0mJ75MiRatiwoS655BKVKFFCnTt31m+//XbBx33wwQeqVq2acufOrVq1aumLL77wWW9mGjZsmEqVKqU8efKoTZs22rRpU6BeBgAAAAAAPkJabH/77bfq27evli5dqq+++kqnT5/Wddddp/j4+HM+5scff1S3bt3Uu3dvrVq1Sp07d1bnzp21du1a7zYvvPCCXnnlFU2YMEHLli1Tvnz5FBMTo1OnTgXjZQEAAAAAsjmPmVmoQ6T6448/VKJECX377be6+uqr092ma9euio+P12effeZddtVVV6lu3bqaMGGCzEylS5fWv//9bz366KOSpKNHj6pkyZKaOnWqbr/99gvmOHbsmAoWLKijR4+qQIEC/nlxAAAAAIB0ZcUaLKzO2T569KgkqUiRIufcZsmSJWrTpo3PspiYGC1ZskSStG3bNsXFxflsU7BgQTVu3Ni7DQAAAAAAgZQj1AFSpaSkaMCAAWrWrJlq1qx5zu3i4uJUsmRJn2UlS5ZUXFycd33qsnNtc6aEhAQlJCR47x87duyiXgMAAAAAAFIYFdt9+/bV2rVr9f333wf9uUeOHKkRI0acd5vrLrsrSGkubP6Ody64TYfafYOQ5MK+WDM+1BEAAAAAIOjC4jDyfv366bPPPtPChQt16aWXnnfb6Oho7du3z2fZvn37FB0d7V2fuuxc25xpyJAhOnr0qPe2c+fOi30pAAAAAACEttg2M/Xr10+zZ8/WN998o/Lly1/wMU2aNNGCBQt8ln311Vdq0qSJJKl8+fKKjo722ebYsWNatmyZd5szRUVFqUCBAj43AAAAAAAuVkgPI+/bt6/ee+89ffzxx7rkkku851QXLFhQefLkkSR1795dZcqU0ciRIyVJ/fv31zXXXKOXXnpJ119/vaZPn66ffvpJkyZNkiR5PB4NGDBA//nPf1S5cmWVL19eTz75pEqXLq3OnTuH5HUCAAAAALKXkBbbr732miSpZcuWPsvffPNN9ezZU5K0Y8cORUT8/QV806ZN9d5772no0KF6/PHHVblyZc2ZM8dnUrVBgwYpPj5e9913n44cOaLmzZtr7ty5yp07d8BfEwAAAAAAYXWd7XCR3jXemCDt4jBBGgAAAIAL4TrbAAAAAADggii2AQAAAADwM4ptAAAAAAD8jGIbAAAAAAA/o9gGAAAAAMDPKLYBAAAAAPAzim0AAAAAAPyMYhsAAAAAAD+j2AYAAAAAwM8otgEAAAAA8DOKbQAAAAAA/IxiGwAAAAAAP6PYBgAAAADAzyi2AQAAAADwM4ptAAAAAAD8jGIbAAAAAAA/o9gGAAAAAMDPKLYBAAAAAPAzim0AAAAAAPyMYhsAAAAAAD+j2AYAAAAAwM8otgEAAAAA8DOKbQAAAAAA/IxiGwAAAAAAP6PYBgAAAADAzyi2AQAAAADwM4ptAAAAAAD8jGIbAAAAAAA/o9gGAAAAAMDPKLYBAAAAAPAzim0AAAAAAPyMYhsAAAAAAD+j2AYAAAAAwM8otgEAAAAA8DOKbQAAAAAA/CykxfZ3332nG2+8UaVLl5bH49GcOXPOu33Pnj3l8XjOutWoUcO7zVNPPXXW+mrVqgX4lQAAAAAA8LeQFtvx8fGqU6eOxo8fn6HtX375Ze3du9d727lzp4oUKaJbb73VZ7saNWr4bPf9998HIj4AAAAAAOnKEconb9++vdq3b5/h7QsWLKiCBQt678+ZM0eHDx9Wr169fLbLkSOHoqOj/ZYTAAAAAIDMcPqc7SlTpqhNmza6/PLLfZZv2rRJpUuXVoUKFXTnnXdqx44d591PQkKCjh075nMDAAAAAOBiOVts79mzR19++aXuvfden+WNGzfW1KlTNXfuXL322mvatm2bWrRooT///POc+xo5cqT3W/OCBQuqbNmygY4PAAAAAMjCnC22p02bpkKFCqlz584+y9u3b69bb71VtWvXVkxMjL744gsdOXJEM2fOPOe+hgwZoqNHj3pvO3fuDHB6AAAAAEBWFtJzti+WmemNN97Q3XffrVy5cp1320KFCqlKlSravHnzObeJiopSVFSUv2MCAAAAALIpJ7/Z/vbbb7V582b17t37gtseP35cW7ZsUalSpYKQDAAAAACAEBfbx48f1+rVq7V69WpJ0rZt27R69WrvhGZDhgxR9+7dz3rclClT1LhxY9WsWfOsdY8++qi+/fZbbd++XT/++KNuuukmRUZGqlu3bgF9LQAAAAAApArpYeQ//fSTWrVq5b0fGxsrSerRo4emTp2qvXv3njWT+NGjR/XRRx/p5ZdfTnefu3btUrdu3XTw4EEVL15czZs319KlS1W8ePHAvRAAAAAAANIIabHdsmVLmdk510+dOvWsZQULFtSJEyfO+Zjp06f7IxoAAAAAABfNyXO2AQAAAAAIZxTbAAAAAAD4GcU2AAAAAAB+RrENAAAAAICfUWwDAAAAAOBnFNsAAAAAAPgZxTYAAAAAAH5GsQ0AAAAAgJ9RbAMAAAAA4GcU2wAAAAAA+BnFNgAAAAAAfkaxDQAAAACAn1FsAwAAAADgZxTbAAAAAAD4GcU2AAAAAAB+RrENAAAAAICfUWwDAAAAAOBnFNsAAAAAAPgZxTYAAAAAAH5GsQ0AAAAAgJ9RbAMAAAAA4GcU2wAAAAAA+BnFNgAAAAAAfkaxDQAAAACAn1FsAwAAAADgZxTbAAAAAAD4GcU2AAAAAAB+RrENAAAAAICfUWwDAAAAAOBnFNsAAAAAAPgZxTYAAAAAAH5GsQ0AAAAAgJ9RbAMAAAAA4GcU2wAAAAAA+FlIi+3vvvtON954o0qXLi2Px6M5c+acd/tFixbJ4/GcdYuLi/PZbvz48SpXrpxy586txo0ba/ny5QF8FQAAAAAA+AppsR0fH686depo/PjxmXrcb7/9pr1793pvJUqU8K6bMWOGYmNjNXz4cK1cuVJ16tRRTEyM9u/f7+/4AAAAAACkK0con7x9+/Zq3759ph9XokQJFSpUKN11o0aNUp8+fdSrVy9J0oQJE/T555/rjTfe0ODBg/9JXAAAAAAAMsTJc7br1q2rUqVKqW3btvrhhx+8yxMTE/Xzzz+rTZs23mURERFq06aNlixZEoqoAAAAAIBsyKliu1SpUpowYYI++ugjffTRRypbtqxatmyplStXSpIOHDig5ORklSxZ0udxJUuWPOu87rQSEhJ07NgxnxsAAAAAABcrpIeRZ1bVqlVVtWpV7/2mTZtqy5YtGj16tN5+++2L3u/IkSM1YsQIf0QEAAAAAMCtb7bT06hRI23evFmSVKxYMUVGRmrfvn0+2+zbt0/R0dHn3MeQIUN09OhR723nzp0BzQwAAAAAyNqcL7ZXr16tUqVKSZJy5cql+vXra8GCBd71KSkpWrBggZo0aXLOfURFRalAgQI+NwAAAAAALlZIDyM/fvy491tpSdq2bZtWr16tIkWK6LLLLtOQIUO0e/duvfXWW5KkMWPGqHz58qpRo4ZOnTqlyZMn65tvvtH8+fO9+4iNjVWPHj3UoEEDNWrUSGPGjFF8fLx3dnIAAAAAAAItpMX2Tz/9pFatWnnvx8bGSpJ69OihqVOnau/evdqxY4d3fWJiov79739r9+7dyps3r2rXrq2vv/7aZx9du3bVH3/8oWHDhikuLk5169bV3Llzz5o0DQAAAACAQPGYmYU6RLg5duyYChYsqKNHj3oPKb/usrtCnOpv83e8c8FtOtTuG4QkF/bFmvGhjgAAAAAgzKVXg7nO+XO2AQAAAAAINxTbAAAAAAD4GcU2AAAAAAB+RrENAAAAAICfUWwDAAAAAOBnFNsAAAAAAPgZxTYAAAAAAH5GsQ0AAAAAgJ9RbAMAAAAA4GcU2wAAAAAA+BnFNgAAAAAAfpbpYnvnzp3atWuX9/7y5cs1YMAATZo0ya/BAAAAAABwVaaL7TvuuEMLFy6UJMXFxalt27Zavny5nnjiCT399NN+DwgAAAAAgGsyXWyvXbtWjRo1kiTNnDlTNWvW1I8//qh3331XU6dO9Xc+AAAAAACck+li+/Tp04qKipIkff311+rYsaMkqVq1atq7d69/0wEAAAAA4KBMF9s1atTQhAkTtHjxYn311Vdq166dJGnPnj0qWrSo3wMCAAAAAOCaTBfbzz//vCZOnKiWLVuqW7duqlOnjiTpk08+8R5eDgAAAABAdpYjMxubmSpUqKAdO3YoKSlJhQsX9q677777lDdvXr8HBAAAAADANZn6ZtvMVKlSJcXFxfkU2pJUrlw5lShRwq/hAAAAAABwUaaK7YiICFWuXFkHDx4MVB4AAAAAAJyX6XO2n3vuOQ0cOFBr164NRB4AAAAAAJyXqXO2Jal79+46ceKE6tSpo1y5cilPnjw+6w8dOuS3cAAAAAAAuCjTxfaYMWMCEAMAAAAAgKwj08V2jx49ApEDAAAAAIAsI9PnbEvSli1bNHToUHXr1k379++XJH355Zdat26dX8MBAAAAAOCiTBfb3377rWrVqqVly5Zp1qxZOn78uCTpl19+0fDhw/0eEAAAAAAA12S62B48eLD+85//6KuvvlKuXLm8y1u3bq2lS5f6NRwAAAAAAC7KdLH966+/6qabbjpreYkSJXTgwAG/hAIAAAAAwGWZLrYLFSqkvXv3nrV81apVKlOmjF9CAQAAAADgskwX27fffrsee+wxxcXFyePxKCUlRT/88IMeffRRde/ePRAZAQAAAABwSqaL7WeffVbVqlVT2bJldfz4cVWvXl1XX321mjZtqqFDhwYiIwAAAAAATsn0dbZz5cql119/XU8++aTWrl2r48eP68orr1TlypUDkQ8AAAAAAOdkuthOddlll+myyy7zZxYAAAAAALKEDBXbsbGxeuaZZ5QvXz7Fxsaed9tRo0b5JRgAAAAAAK7KULG9atUqnT592vv/c/F4PP5JBQAAAACAwzI0QdrChQtVqFAh7//Pdfvmm28y9eTfffedbrzxRpUuXVoej0dz5sw57/azZs1S27ZtVbx4cRUoUEBNmjTRvHnzfLZ56qmn5PF4fG7VqlXLVC4AAAAAAP6JTM9G7k/x8fGqU6eOxo8fn6Htv/vuO7Vt21ZffPGFfv75Z7Vq1Uo33njjWd+216hRQ3v37vXevv/++0DEBwAAAAAgXRk6jLxLly4Z3uGsWbMyvG379u3Vvn37DG8/ZswYn/vPPvusPv74Y3366ae68sorvctz5Mih6OjoDO8XAAAAAAB/ylCxXbBgwUDnuCgpKSn6888/VaRIEZ/lmzZtUunSpZU7d241adJEI0eOPO/M6QkJCUpISPDeP3bsWMAyAwAAAACyvgwV22+++Wagc1yUF198UcePH9dtt93mXda4cWNNnTpVVatW1d69ezVixAi1aNFCa9eu1SWXXJLufkaOHKkRI0YEKzYAAAAAIIvL9DnbrVu31pEjR85afuzYMbVu3dofmTLkvffe04gRIzRz5kyVKFHCu7x9+/a69dZbVbt2bcXExOiLL77QkSNHNHPmzHPua8iQITp69Kj3tnPnzmC8BAAAAABAFpWhb7bTWrRokRITE89afurUKS1evNgvoS5k+vTpuvfee/XBBx+oTZs25922UKFCqlKlijZv3nzObaKiohQVFeXvmAAAAACAbCrDxfaaNWu8/1+/fr3i4uK895OTkzV37lyVKVPGv+nS8f777+uee+7R9OnTdf31119w++PHj2vLli26++67A54NAAAAAAApE8V23bp1vdetTu9w8Tx58mjs2LGZevLjx4/7fOO8bds2rV69WkWKFNFll12mIUOGaPfu3Xrrrbck/XXoeI8ePfTyyy+rcePG3oI/T5483kncHn30Ud144426/PLLtWfPHg0fPlyRkZHq1q1bprIBAAAAAHCxMlxsb9u2TWamChUqaPny5SpevLh3Xa5cuVSiRAlFRkZm6sl/+ukntWrVyns/NjZWktSjRw9NnTpVe/fu1Y4dO7zrJ02apKSkJPXt21d9+/b1Lk/dXpJ27dqlbt266eDBgypevLiaN2+upUuX+uQFAAAAACCQMlxsX3755ZL+utyWv7Rs2VJmds71qQV0qkWLFl1wn9OnT/+HqQAAAAAA+GcyPUFa6iHd59K9e/eLDgMAAAAAQFaQ6WK7f//+PvdPnz6tEydOKFeuXMqbNy/FNgAAAAAg28v0dbYPHz7sczt+/Lh+++03NW/eXO+//34gMgIAAAAA4JRMF9vpqVy5sp577rmzvvUGAAAAACA78kuxLUk5cuTQnj17/LU7AAAAAACclelztj/55BOf+2amvXv3aty4cWrWrJnfggEAAAAA4KpMF9udO3f2ue/xeFS8eHG1bt1aL730kr9yAQAAAADgrEwX2/68zjYAAAAAAFnRRZ+zfeDAAR04cMCfWQAAAAAAyBIyVWwfOXJEffv2VbFixVSyZEmVLFlSxYoVU79+/XTkyJEARQQAAAAAwC0ZPoz80KFDatKkiXbv3q0777xTV1xxhSRp/fr1mjp1qhYsWKAff/xRhQsXDlhYAAAAAABckOFi++mnn1auXLm0ZcsWlSxZ8qx11113nZ5++mmNHj3a7yEBAAAAAHBJhg8jnzNnjl588cWzCm1Jio6O1gsvvKDZs2f7NRwAAAAAAC7KcLG9d+9e1ahR45zra9asqbi4OL+EAgAAAADAZRkutosVK6bt27efc/22bdtUpEgRf2QCAAAAAMBpGS62Y2Ji9MQTTygxMfGsdQkJCXryySfVrl07v4YDAAAAAMBFmZogrUGDBqpcubL69u2ratWqycy0YcMGvfrqq0pISNDbb78dyKwAAAAAADghw8X2pZdeqiVLluihhx7SkCFDZGaSJI/Ho7Zt22rcuHEqW7ZswIICAAAAAOCKDBfbklS+fHl9+eWXOnz4sDZt2iRJqlSpEudqAwAAAACQRqaK7VSFCxdWo0aN/J0FAAAAAIAsIcMTpAEAAAAAgIyh2AYAAAAAwM8otgEAAAAA8LMMFdv16tXT4cOHJf11CbATJ04ENBQAAAAAAC7LULG9YcMGxcfHS5JGjBih48ePBzQUAAAAAAAuy9Bs5HXr1lWvXr3UvHlzmZlefPFF5c+fP91thw0b5teAAAAAAAC4JkPF9tSpUzV8+HB99tln8ng8+vLLL5Ujx9kP9Xg8FNsAAAAAgGwvQ8V21apVNX36dElSRESEFixYoBIlSgQ0GAAAAAAArspQsZ1WSkpKIHIAAAAAAJBlZLrYlqQtW7ZozJgx2rBhgySpevXq6t+/vypWrOjXcAAAAAAAuCjT19meN2+eqlevruXLl6t27dqqXbu2li1bpho1auirr74KREYAAAAAAJyS6W+2Bw8erEceeUTPPffcWcsfe+wxtW3b1m/hAAAAAABwUaa/2d6wYYN69+591vJ77rlH69ev90soAAAAAABcluliu3jx4lq9evVZy1evXs0M5QAAAAAA6CIOI+/Tp4/uu+8+bd26VU2bNpUk/fDDD3r++ecVGxvr94AAAAAAALgm099sP/nkkxo2bJjGjh2ra665Rtdcc43GjRunp556SkOHDs3Uvr777jvdeOONKl26tDwej+bMmXPBxyxatEj16tVTVFSUKlWqpKlTp561zfjx41WuXDnlzp1bjRs31vLlyzOVCwAAAACAfyLTxbbH49EjjzyiXbt26ejRozp69Kh27dql/v37y+PxZGpf8fHxqlOnjsaPH5+h7bdt26brr79erVq10urVqzVgwADde++9mjdvnnebGTNmKDY2VsOHD9fKlStVp04dxcTEaP/+/ZnKBgAAAADAxbqo62ynuuSSS/7Rk7dv317t27fP8PYTJkxQ+fLl9dJLL0mSrrjiCn3//fcaPXq0YmJiJEmjRo1Snz591KtXL+9jPv/8c73xxhsaPHjwP8oLAAAAAEBGZPqb7VBasmSJ2rRp47MsJiZGS5YskSQlJibq559/9tkmIiJCbdq08W6TnoSEBB07dsznBgAAAADAxfpH32wHW1xcnEqWLOmzrGTJkjp27JhOnjypw4cPKzk5Od1tNm7ceM79jhw5UiNGjAhIZlzYDc0HhTqCJOmz71+44DbXdxgWhCQZ8/kXT593fcwt4dOn5304/ILbtO7+TBCSXNg3bz15wW1a3B8eWRdPvHDWJv3DI6skLXn5/HkbPnb+Ph1MK56/8Hv9yuHh8R5bNeLC7686zz0V+CAZ8Mvgpy64Td1XLvx6gmX1w+f/GV81JXNz1QTS0t7/ueA2100fEoQkFzb/9pEX3OaOTx8NQpILe+/GFy+4TezCfkFIkjGjWo077/pnF3cPUpILe7zFWxfc5vUlHYOQ5ML6NPnkgtt8vKxFEJJcWKfGiy+4zdIV9YOQJGOuavhzqCMElFPfbAfKkCFDvOefHz16VDt37gx1JAAAAACAwzJVbJ8+fVrXXnutNm3aFKg85xUdHa19+/b5LNu3b58KFCigPHnyqFixYoqMjEx3m+jo6HPuNyoqSgUKFPC5AQAAAABwsTJVbOfMmVNr1qwJVJYLatKkiRYsWOCz7KuvvlKTJk0kSbly5VL9+vV9tklJSdGCBQu82wAAAAAAEGiZPoz8rrvu0pQpU/zy5MePH9fq1au1evVqSX9d2mv16tXasWOHpL8O7+7e/e9zSx544AFt3bpVgwYN0saNG/Xqq69q5syZeuSRR7zbxMbG6vXXX9e0adO0YcMGPfjgg4qPj/fOTg4AAAAAQKBleoK0pKQkvfHGG/r6669Vv3595cuXz2f9qFGjMryvn376Sa1atfLej42NlST16NFDU6dO1d69e72FtySVL19en3/+uR555BG9/PLLuvTSSzV58mTvZb8kqWvXrvrjjz80bNgwxcXFqW7dupo7d+5Zk6YBAAAAABAomS62165dq3r16kmS/u///s9nncfjydS+WrZsKTM75/qpU6em+5hVq1add7/9+vVTv37hMzMkAAAAACB7yXSxvXDhwkDkAAAAAAAgy7joS39t3rxZ8+bN08mTJyXpvN9QAwAAAACQnWS62D548KCuvfZaValSRR06dNDevXslSb1799a///1vvwcEAAAAAMA1mS62H3nkEeXMmVM7duxQ3rx5vcu7du2quXPn+jUcAAAAAAAuyvQ52/Pnz9e8efN06aWX+iyvXLmyfv/9d78FAwAAAADAVZn+Zjs+Pt7nG+1Uhw4dUlRUlF9CAQAAAADgskwX2y1atNBbb73lve/xeJSSkqIXXnjB55rZAAAAAABkV5k+jPyFF17Qtddeq59++kmJiYkaNGiQ1q1bp0OHDumHH34IREYAAAAAAJyS6W+2a9asqf/7v/9T8+bN1alTJ8XHx6tLly5atWqVKlasGIiMAAAAAAA4JdPfbEtSwYIF9cQTT/g7CwAAAAAAWcJFFduHDx/WlClTtGHDBklS9erV1atXLxUpUsSv4QAAAAAAcFGmDyP/7rvvVK5cOb3yyis6fPiwDh8+rFdeeUXly5fXd999F4iMAAAAAAA4JdPfbPft21ddu3bVa6+9psjISElScnKyHnroIfXt21e//vqr30MCAAAAAOCSTH+zvXnzZv373//2FtqSFBkZqdjYWG3evNmv4QAAAAAAcFGmi+169ep5z9VOa8OGDapTp45fQgEAAAAA4LIMHUa+Zs0a7/8ffvhh9e/fX5s3b9ZVV10lSVq6dKnGjx+v5557LjApAQAAAABwSIaK7bp168rj8cjMvMsGDRp01nZ33HGHunbt6r90AAAAAAA4KEPF9rZt2wKdAwAAAACALCNDxfbll18e6BwAAAAAAGQZmb70lyTt2bNH33//vfbv36+UlBSfdQ8//LBfggEAAAAA4KpMF9tTp07V/fffr1y5cqlo0aLyeDzedR6Ph2IbAAAAAJDtZbrYfvLJJzVs2DANGTJEERGZvnIYAAAAAABZXqar5RMnTuj222+n0AYAAAAA4BwyXTH37t1bH3zwQSCyAAAAAACQJWT6MPKRI0fqhhtu0Ny5c1WrVi3lzJnTZ/2oUaP8Fg4AAAAAABddVLE9b948Va1aVZLOmiANAAAAAIDsLtPF9ksvvaQ33nhDPXv2DEAcAAAAAADcl+lztqOiotSsWbNAZAEAAAAAIEvIdLHdv39/jR07NhBZAAAAAADIEjJ9GPny5cv1zTff6LPPPlONGjXOmiBt1qxZfgsHAAAAAICLMl1sFypUSF26dAlEFgAAAAAAsoRMF9tvvvlmIHIAAAAAAJBlZPqcbQAAAAAAcH6Z/ma7fPny572e9tatW/9RIAAAAAAAXJfpb7YHDBig/v37e28PPfSQmjRpoqNHj+q+++67qBDjx49XuXLllDt3bjVu3FjLly8/57YtW7aUx+M563b99dd7t+nZs+dZ69u1a3dR2QAAAAAAyKxMf7Pdv3//dJePHz9eP/30U6YDzJgxQ7GxsZowYYIaN26sMWPGKCYmRr/99ptKlChx1vazZs1SYmKi9/7BgwdVp04d3XrrrT7btWvXzuf88qioqExnAwAAAADgYvjtnO327dvro48+yvTjRo0apT59+qhXr16qXr26JkyYoLx58+qNN95Id/siRYooOjrae/vqq6+UN2/es4rtqKgon+0KFy58Ua8LAAAAAIDM8lux/eGHH6pIkSKZekxiYqJ+/vlntWnT5u9AERFq06aNlixZkqF9TJkyRbfffrvy5cvns3zRokUqUaKEqlatqgcffFAHDx7MVDYAAAAAAC5Wpg8jv/LKK30mSDMzxcXF6Y8//tCrr76aqX0dOHBAycnJKlmypM/ykiVLauPGjRd8/PLly7V27VpNmTLFZ3m7du3UpUsXlS9fXlu2bNHjjz+u9u3ba8mSJYqMjDxrPwkJCUpISPDeP3bsWKZeBwAAAAAAaWW62O7cubPP/YiICBUvXlwtW7ZUtWrV/JUrQ6ZMmaJatWqpUaNGPstvv/127/9r1aql2rVrq2LFilq0aJGuvfbas/YzcuRIjRgxIuB5AQAAAADZQ6aL7eHDh/vtyYsVK6bIyEjt27fPZ/m+ffsUHR193sfGx8dr+vTpevrppy/4PBUqVFCxYsW0efPmdIvtIUOGKDY21nv/2LFjKlu2bAZfBQAAAAAAvvx2zvbFyJUrl+rXr68FCxZ4l6WkpGjBggVq0qTJeR/7wQcfKCEhQXfdddcFn2fXrl06ePCgSpUqle76qKgoFShQwOcGAAAAAMDFynCxHRERocjIyPPecuTI9Bflio2N1euvv65p06Zpw4YNevDBBxUfH69evXpJkrp3764hQ4ac9bgpU6aoc+fOKlq0qM/y48ePa+DAgVq6dKm2b9+uBQsWqFOnTqpUqZJiYmIynQ8AAAAAgMzKcHU8e/bsc65bsmSJXnnlFaWkpGQ6QNeuXfXHH39o2LBhiouLU926dTV37lzvpGk7duxQRITvZwK//fabvv/+e82fP/+s/UVGRmrNmjWaNm2ajhw5otKlS+u6667TM888w7W2AQAAAABBkeFiu1OnTmct++233zR48GB9+umnuvPOOzN0/nR6+vXrp379+qW7btGiRWctq1q1qsws3e3z5MmjefPmXVQOAAAAAAD84aLO2d6zZ4/69OmjWrVqKSkpSatXr9a0adN0+eWX+zsfAAAAAADOyVSxffToUT322GOqVKmS1q1bpwULFujTTz9VzZo1A5UPAAAAAADnZPgw8hdeeEHPP/+8oqOj9f7776d7WDkAAAAAAMhEsT148GDlyZNHlSpV0rRp0zRt2rR0t5s1a5bfwgEAAAAA4KIMF9vdu3eXx+MJZBYAAAAAALKEDBfbU6dODWAMAAAAAACyjouajRwAAAAAAJwbxTYAAAAAAH5GsQ0AAAAAgJ9RbAMAAAAA4GcU2wAAAAAA+BnFNgAAAAAAfkaxDQAAAACAn1FsAwAAAADgZxTbAAAAAAD4GcU2AAAAAAB+RrENAAAAAICfUWwDAAAAAOBnFNsAAAAAAPgZxTYAAAAAAH5GsQ0AAAAAgJ9RbAMAAAAA4GcU2wAAAAAA+BnFNgAAAAAAfkaxDQAAAACAn1FsAwAAAADgZxTbAAAAAAD4GcU2AAAAAAB+RrENAAAAAICfUWwDAAAAAOBnFNsAAAAAAPgZxTYAAAAAAH5GsQ0AAAAAgJ9RbAMAAAAA4GcU2wAAAAAA+FlYFNvjx49XuXLllDt3bjVu3FjLly8/57ZTp06Vx+PxueXOndtnGzPTsGHDVKpUKeXJk0dt2rTRpk2bAv0yAAAAAACQFAbF9owZMxQbG6vhw4dr5cqVqlOnjmJiYrR///5zPqZAgQLau3ev9/b777/7rH/hhRf0yiuvaMKECVq2bJny5cunmJgYnTp1KtAvBwAAAACA0Bfbo0aNUp8+fdSrVy9Vr15dEyZMUN68efXGG2+c8zEej0fR0dHeW8mSJb3rzExjxozR0KFD1alTJ9WuXVtvvfWW9uzZozlz5gThFQEAAAAAsruQFtuJiYn6+eef1aZNG++yiIgItWnTRkuWLDnn444fP67LL79cZcuWVadOnbRu3Trvum3btikuLs5nnwULFlTjxo3Puc+EhAQdO3bM5wYAAAAAwMUKabF94MABJScn+3wzLUklS5ZUXFxcuo+pWrWq3njjDX388cd65513lJKSoqZNm2rXrl2S5H1cZvY5cuRIFSxY0HsrW7bsP31pAAAAAIBsLOSHkWdWkyZN1L17d9WtW1fXXHONZs2apeLFi2vixIkXvc8hQ4bo6NGj3tvOnTv9mBgAAAAAkN2EtNguVqyYIiMjtW/fPp/l+/btU3R0dIb2kTNnTl155ZXavHmzJHkfl5l9RkVFqUCBAj43AAAAAAAuVkiL7Vy5cql+/fpasGCBd1lKSooWLFigJk2aZGgfycnJ+vXXX1WqVClJUvny5RUdHe2zz2PHjmnZsmUZ3icAAAAAAP9EjlAHiI2NVY8ePdSgQQM1atRIY8aMUXx8vHr16iVJ6t69u8qUKaORI0dKkp5++mldddVVqlSpko4cOaL//e9/+v3333XvvfdK+mum8gEDBug///mPKleurPLly+vJJ59U6dKl1blz51C9TAAAAABANhLyYrtr1676448/NGzYMMXFxalu3bqaO3eud4KzHTt2KCLi7y/gDx8+rD59+iguLk6FCxdW/fr19eOPP6p69erebQYNGqT4+Hjdd999OnLkiJo3b665c+cqd+7cQX99AAAAAIDsJ+TFtiT169dP/fr1S3fdokWLfO6PHj1ao0ePPu/+PB6Pnn76aT399NP+iggAAAAAQIY5Nxs5AAAAAADhjmIbAAAAAAA/o9gGAAAAAMDPKLYBAAAAAPAzim0AAAAAAPyMYhsAAAAAAD+j2AYAAAAAwM8otgEAAAAA8DOKbQAAAAAA/IxiGwAAAAAAP6PYBgAAAADAzyi2AQAAAADwM4ptAAAAAAD8jGIbAAAAAAA/o9gGAAAAAMDPKLYBAAAAAPAzim0AAAAAAPyMYhsAAAAAAD+j2AYAAAAAwM8otgEAAAAA8DOKbQAAAAAA/IxiGwAAAAAAP6PYBgAAAADAzyi2AQAAAADwM4ptAAAAAAD8jGIbAAAAAAA/o9gGAAAAAMDPKLYBAAAAAPAzim0AAAAAAPyMYhsAAAAAAD+j2AYAAAAAwM8otgEAAAAA8DOKbQAAAAAA/IxiGwAAAAAAPwuLYnv8+PEqV66ccufOrcaNG2v58uXn3Pb1119XixYtVLhwYRUuXFht2rQ5a/uePXvK4/H43Nq1axfolwEAAAAAgKQwKLZnzJih2NhYDR8+XCtXrlSdOnUUExOj/fv3p7v9okWL1K1bNy1cuFBLlixR2bJldd1112n37t0+27Vr10579+713t5///1gvBwAAAAAAEJfbI8aNUp9+vRRr169VL16dU2YMEF58+bVG2+8ke727777rh566CHVrVtX1apV0+TJk5WSkqIFCxb4bBcVFaXo6GjvrXDhwsF4OQAAAAAAhLbYTkxM1M8//6w2bdp4l0VERKhNmzZasmRJhvZx4sQJnT59WkWKFPFZvmjRIpUoUUJVq1bVgw8+qIMHD/o1OwAAAAAA55IjlE9+4MABJScnq2TJkj7LS5YsqY0bN2ZoH4899phKly7tU7C3a9dOXbp0Ufny5bVlyxY9/vjjat++vZYsWaLIyMiz9pGQkKCEhATv/WPHjl3kKwIAAAAAIMTF9j/13HPPafr06Vq0aJFy587tXX777bd7/1+rVi3Vrl1bFStW1KJFi3TttdeetZ+RI0dqxIgRQckMAAAAAMj6QnoYebFixRQZGal9+/b5LN+3b5+io6PP+9gXX3xRzz33nObPn6/atWufd9sKFSqoWLFi2rx5c7rrhwwZoqNHj3pvO3fuzNwLAQAAAAAgjZAW27ly5VL9+vV9JjdLneysSZMm53zcCy+8oGeeeUZz585VgwYNLvg8u3bt0sGDB1WqVKl010dFRalAgQI+NwAAAAAALlbIZyOPjY3V66+/rmnTpmnDhg168MEHFR8fr169ekmSunfvriFDhni3f/755/Xkk0/qjTfeULly5RQXF6e4uDgdP35cknT8+HENHDhQS5cu1fbt27VgwQJ16tRJlSpVUkxMTEheIwAAAAAgewn5Odtdu3bVH3/8oWHDhikuLk5169bV3LlzvZOm7dixQxERf38m8NprrykxMVG33HKLz36GDx+up556SpGRkVqzZo2mTZumI0eOqHTp0rruuuv0zDPPKCoqKqivDQAAAACQPYW82Jakfv36qV+/fumuW7Rokc/97du3n3dfefLk0bx58/yUDAAAAACAzAv5YeQAAAAAAGQ1FNsAAAAAAPgZxTYAAAAAAH5GsQ0AAAAAgJ9RbAMAAAAA4GcU2wAAAAAA+BnFNgAAAAAAfkaxDQAAAACAn1FsAwAAAADgZxTbAAAAAAD4GcU2AAAAAAB+RrENAAAAAICfUWwDAAAAAOBnFNsAAAAAAPgZxTYAAAAAAH5GsQ0AAAAAgJ9RbAMAAAAA4GcU2wAAAAAA+BnFNgAAAAAAfkaxDQAAAACAn1FsAwAAAADgZxTbAAAAAAD4GcU2AAAAAAB+RrENAAAAAICfUWwDAAAAAOBnFNsAAAAAAPgZxTYAAAAAAH5GsQ0AAAAAgJ9RbAMAAAAA4GcU2wAAAAAA+BnFNgAAAAAAfkaxDQAAAACAn1FsAwAAAADgZxTbAAAAAAD4GcU2AAAAAAB+FhbF9vjx41WuXDnlzp1bjRs31vLly8+7/QcffKBq1aopd+7cqlWrlr744guf9WamYcOGqVSpUsqTJ4/atGmjTZs2BfIlAAAAAADgFfJie8aMGYqNjdXw4cO1cuVK1alTRzExMdq/f3+62//444/q1q2bevfurVWrVqlz587q3Lmz1q5d693mhRde0CuvvKIJEyZo2bJlypcvn2JiYnTq1KlgvSwAAAAAQDYW8mJ71KhR6tOnj3r16qXq1atrwoQJyps3r9544410t3/55ZfVrl07DRw4UFdccYWeeeYZ1atXT+PGjZP017faY8aM0dChQ9WpUyfVrl1bb731lvbs2aM5c+YE8ZUBAAAAALKrkBbbiYmJ+vnnn9WmTRvvsoiICLVp00ZLlixJ9zFLlizx2V6SYmJivNtv27ZNcXFxPtsULFhQjRs3Puc+AQAAAADwpxyhfPIDBw4oOTlZJUuW9FlesmRJbdy4Md3HxMXFpbt9XFycd33qsnNtc6aEhAQlJCR47x89elSSdOzYMe+ypJTTGXlJQZE217mcTk4MQpILy1DWpIQLbhMMGcp6OjyyShfOm3Q6fE6byEjbJiWGR94slzUhPLJKF86b7FBWKXzyZijrqfAYu1zKKmVgnD3pTlZJSjoRHnkz9PvWoawJ8eHxN5d04bynHMoqSSfjw+Pv74xkPRGfFIQkF5aRrPHHk4OQJGPS5k39v5mFKo7/WQjt3r3bJNmPP/7os3zgwIHWqFGjdB+TM2dOe++993yWjR8/3kqUKGFmZj/88INJsj179vhsc+utt9ptt92W7j6HDx9ukrhx48aNGzdu3Lhx48aNWwhvO3fuvNjyMuyE9JvtYsWKKTIyUvv27fNZvm/fPkVHR6f7mOjo6PNun/rvvn37VKpUKZ9t6tatm+4+hwwZotjYWO/9lJQUHTp0SEWLFpXH48n06zqXY8eOqWzZstq5c6cKFCjgt/0GAlkDw6Wsklt5yRoYLmWV3MpL1sBwKavkVl6yBoZLWSW38pI1MAKV1cz0559/qnTp0n7bZ6iFtNjOlSuX6tevrwULFqhz586S/ip0FyxYoH79+qX7mCZNmmjBggUaMGCAd9lXX32lJk2aSJLKly+v6OhoLViwwFtcHzt2TMuWLdODDz6Y7j6joqIUFRXls6xQoUL/6LWdT4ECBcL+TZSKrIHhUlbJrbxkDQyXskpu5SVrYLiUVXIrL1kDw6Wsklt5yRoYgchasGBBv+4v1EJabEtSbGysevTooQYNGqhRo0YaM2aM4uPj1atXL0lS9+7dVaZMGY0cOVKS1L9/f11zzTV66aWXdP3112v69On66aefNGnSJEmSx+PRgAED9J///EeVK1dW+fLl9eSTT6p06dLegh4AAAAAgEAKebHdtWtX/fHHHxo2bJji4uJUt25dzZ071zvB2Y4dOxQR8fek6U2bNtV7772noUOH6vHHH1flypU1Z84c1axZ07vNoEGDFB8fr/vuu09HjhxR8+bNNXfuXOXOnTvorw8AAAAAkP2EvNiWpH79+p3zsPFFixadtezWW2/Vrbfees79eTwePf3003r66af9FdEvoqKiNHz48LMOWQ9HZA0Ml7JKbuUla2C4lFVyKy9ZA8OlrJJbeckaGC5lldzKS9bAcClrqHnMstLc6gAAAAAAhF7EhTcBAAAAAACZQbENAAAAAICfUWwDAAAAAOBnFNsAAAAAAPgZxTYAAACAgGE+ZmRXFNsAsqWUlJRQR7hoLvzR4kJGF33zzTeaN29eqGNkWGJioiS3+oNLWV1CuwZeOLexx+PxuR/OWRFYrvzs/fV3YlhcZxvuW7VqlXbs2KEcOXKocePGKlasWKgjnVd8fLzy5csX6hjnlJycrMjISJnZWb+gws3YsWN1+PBhDRs2LNRRLujjjz9WQkKCbrvtNkVERDjRvql+/fVXHTt2TLly5dIVV1yh/PnzS1LYvIatW7fq2LFjioqKUrVq1eTxeMIm2/msWbNGf/75pySpWrVqKlq0qKTwade03nzzTfXu3Vt9+/ZVTEyMUlJSFBERvp+Zf/DBB1q6dKliY2NVpkyZsGzTVFu3btWJEycUERGh6tWrhzrOea1YsUK7d+9Wrly5VL16dZUrV05SePbZcB+3JOnIkSM6ceKESpQooRw5/vqzOJzyncvy5csVFxcnSapataqqVq0atuPuF198oS+//FKnT59Ww4YN1bt377DLmGrJkiVav369PB6PGjVqpJo1a4Y60nlt2LBBhw4dksfj0ZVXXqk8efKEOtI5HThwQMWKFQvbfrps2TLt2LFDx48f1913360cOXL4J6cB/9Abb7xhl156qdWvX988Ho899NBDlpKSEupY5/T+++9b7969bfPmzaGOkq5PP/3UnnnmGTt69KiZWVi35aRJk8zj8dj06dNDHeWC1q9fbx6Px6Kjo23q1Kne5eHcvqkmT55sxYsXt2rVqpnH47GOHTva22+/7V0f6tcwdepUq127tkVHR9uVV15pzz//fEjzZNTkyZMtOjramjZtapdccom1a9fOJk+e7F0f6nZNa8KECZYrVy675ZZbrEiRIrZ+/fpQRzqvvXv3WvHixa169eo2ePBg2717d6gjndNbb71l1atXtwoVKlhkZKQ9/fTTZhZeP/9UU6ZMscKFC9uVV15pUVFR1qxZM3vhhRe868Mpc7iPW2Zm7777rrVs2dKio6MtJibG3nvvvVBHypDJkydb0aJFrW3btlaqVClr1KiRPfPMM9714dC2qaZMmWKFChWy7t2727XXXmt16tSxb7/9NtSx0jVlyhQrWbKkXXPNNVa4cGG77bbbbP/+/aGOdU5vvvmmVahQwapWrWoej8ceeOAB27dvX6hjpeu9996zDh062MKFC73Lwq2fli1b1mrVqmVly5a19u3bW3Jysl/2TbGNf+Sdd96xAgUK2IwZM+zUqVM2a9Ys83g8tnPnTp/twuUN9cknn1iePHnM4/HYnXfeab///nuoI/lIbb9KlSrZc889F9YF98SJEy1nzpw2c+ZMMzNLTEy0hIQEn238NVD5w7Zt26xChQr24IMPWvXq1Z0puJctW2aFCxe2d955x/744w9bvHix3XrrrVanTh178cUXQx3Ppk2bZvny5bNp06bZihUr7N5777Wrr77apy+EUz9I9fXXX1uJEiVs5syZFh8fbytXrrRmzZpZnjx5fIqXcDBx4kSLjIy02bNn26FDh6x27dregjAc29bM7MiRI1anTh3r0KGD1atXzwYNGmTbt2/32SYc3nfvv/++5c+f3959911bs2aNvfXWW+bxeOzrr78OdbSzrFmzxkqUKGFvvfWWnTx50tasWWOPPvqoXXrppfbvf//bu104tGu4j1tmZm+//bYVKFDARo8ebe+//761b9/err76ajt48GCoo53X4sWLrWTJkt7fvZs2bbK77rrLPB6PDRw40LtdOPSDpUuXWqlSpWzGjBlm9tfv4erVq9vixYt9tguHrO+9954VLFjQ264LFy60XLly2YYNG3y2C4esZmYzZ860ggUL2owZM2zPnj02b948y5MnT1h+YPT5559boUKFLF++fNa5c2efn384tOenn35qBQoUsDlz5tjevXttxowZVrt2bTt8+LB3m3+Sk2IbF23Dhg3WvHlzmzBhgnfZkSNH7LrrrrM33njDpk2bZkuWLAlhQl/79++3u+66y4YOHWoLFy60vHnz2m233XbWH4ChsmXLFmvWrJkNHjzYHnroIatfv77997//DcuC+7vvvjOPx2NvvfWWmZmtXbvWevfubQ0aNLDrr7/eRo4caUlJSWYWXrnbtm1rM2fOtAcffNAqV65sH330kZmZ/fDDD3b69OkQp0vfe++9Z3Xq1PEpXv/v//7PYmNjrUaNGj7vv2BbsWKFVahQwefbqh9//NFuvfVW+/bbb2358uXefhBuReGIESPslltuMTPzZnzllVesaNGiVqFCBXvllVdCGc9r6tSp5vF4bPbs2d5l9957r1WsWDEs32Nmf+fp3bu3rVq1ykaPHm21a9e24cOH244dO8Km2Nq0aZM1b97cXn31VTP7O3dMTIw99thjPsvCwbx586xKlSo+fwDu27fPXn75ZYuOjrannnoqdOHOEM7jlpnZ6tWrrWbNmjZp0iTvsgMHDtgll1xi77//fgiTXdjYsWMtJibGzP4eV2fMmGElS5a0fPny2eOPPx7KeD6mTp1qDRs29I5VZmbNmjWzbt26WdeuXW3o0KHe5aF8r23fvt3atm1rL7/8sk+eVq1a2bPPPmv/+9//fMbgUNuxY4fFxMR4x9LUftC9e3e78847zSx8xq4jR45Ynz59bMCAAbZ48WKrXr26XX/99WFTcCcnJ1tsbKz961//8i7bvXu3NW/e3CZPnmz/+9//LC4uzswuPmf4nuyFsJc3b1717NlTN954o3fZ3XffrRUrVmjGjBl6/vnn9cgjj+jLL78MYcq/RUVFqUOHDrr22mvVsmVLLV68WJ999pkGDRqk33//PdTxVKRIEcXExOjmm2/W+PHj1axZM82ePVvjxo3TsWPHvOe4pLIQTjCRP39+XXHFFZo6dapWrFihLl266M8//1S7du1UsGBBzZw5Uw8++KBSUlLC4pycpKQknTp1SqdPn9Zll12mQYMG6cYbb9Tjjz+uggULavLkyWe1b6ilZrnkkkt06NAh/fbbb951lStXVr9+/dSsWTPNmDFDGzduDEnGyMhIxcbGqm3btt5lzzzzjBYvXqxu3bqpT58+atWqlZKSksLm3OLUdj148KCOHTsm6a/XIf01mVeLFi3UsmVLzZ49W3v27AlZzlRFihTRZ599ps6dOyspKUmS9MQTTygxMVGjRo2SdPbEQ6GWmicyMlLz5s3TgAEDdMcdd+jjjz9WvXr1NHHiREmhnyTn9OnTSkpKUsOGDSX9nbtEiRLasWOHpNBnTKtIkSI6ePCgfvrpJ++yEiVK6M4771S/fv00a9YsffvttyFM+LdwHrckaefOnapVq5batGkj6a/fEUWLFlW9evWUkJAgSWHz+/ZMJ0+e1MGDB/XHH394x9WDBw+qYcOG+ve//61PPvlEv/zyS4hT/qVAgQLas2eP3nnnHSUmJqpTp07avn27KlasqMsvv1yjR49W//79JYV2HMuXL58eeOABdenSxbusc+fOWrNmjX755Rd98cUXGjx4sN54442QZUwrOTlZuXLlUrNmzST93XYlS5b0nscfLvLmzauuXbuqY8eOat68uT788ENt27ZNzz33nL7//ntJof3ZR0REaPfu3VqyZIl3Wd++fbV582a9/fbbevvtt1WrVi3t2rXr4v9O/AcfBgDeb13NzF599VUrW7asrV271sz++uQt9fDBcPHnn3+a2d+fTq1cudLy5s1rt956q/eQ8iNHjpx1iFOwHD9+3Pv/5ORke/jhh61Bgwb23//+144dO2ZmZocOHQqLbwnXrl1rNWvWNI/HY4MGDbKTJ0+a2V+Hkz///PNWvXp1+/XXX0Oc8i+pP+9HH33U+y3W6tWrrVixYpY/f/6zPs0OJ2vWrLHy5cvbsGHD7NSpU2etK1SokL3zzjshyZaYmGgHDhzw3r/rrrusQoUKtnLlStu5c6fNnz/fqlSp4m3zcDJjxgzzeDw2fvx4W758ub399tsWERFhixcvti1btljevHntu+++C3XMdB0/fty6dOli7du3D3WUdKUeJfLiiy9at27dvMujo6OtYMGC9q9//cv7TUGopT33PTEx0czMBg4caPfee6/Pdnv27AlqrvTs3r3bWrRoYffdd5/t2rXLZ9327dvtiiuuCJujBn799dewHbfM/jrS7YsvvvDeTx33b7jhBhs3bpzPtuH2O2HBggVWsGBBe/zxx23OnDk2efJki4yMtC+//NKOHDliJUuWDJtvYTdv3my9evWywoUL23XXXWfFihXzOSx7zJgxVrly5bNOPQyFEydOeP//1ltvWZkyZbzjw+HDh61jx452xx13hM1RcFu2bPH+PzXT2LFj7aabbvLZLhzm90gdW1OtX7/eqlevbh06dPD+vX3kyBH75JNPQhHPPvnkE6tSpYqVL1/emjZtamXLlrWtW7faiRMnLDk52Ro0aGC33XbbRe+fYhuZcvjwYdu9e/dZb5zUdWee69SpUye75557ghXvLOnlTUlJsZSUFO9hTT///LPlzZvXunbtaitWrLCrr77a7rjjjlBFNrO/B86UlBT717/+ZQ0aNLCRI0fapk2brEWLFjZgwICQ5ku1evVqe+KJJ2zVqlVm9vehTFu3bjWPxxOygfNcBg4c6J3Ar27dutagQQO75557rEaNGmFTEG7ZssVWrVpl69ev97bn+PHjzePx2MSJE30OxzMzu+aaa2zIkCGhiHqW5cuX244dO7z3Dx48aOXLl7dRo0aFMNVfUseCtIe1Pv/881awYEGrWLGiFStWzN59910z++uw8ssuu8x7nmGosp45bqX9d/HixRYREWGzZs0KSca0UvOe+Ufo0qVL7fbbb7fExESrVauWtWzZ0oYMGWINGjSw+++/PyQTD6XXD8z+GrtS32+PPvqo9w+rlJQUu+GGG+y5554Letb0TsGYOXOm5c2b1x5//PGzJkK6+eabrXfv3kHNmCq9Phuu49aZxXPa+y1atLARI0Z479977732xhtvBC1belLbLm0bTp061apVq2YVK1a0yy67zGeS0mrVqtn48eODntMs/T77xx9/2Pbt223ChAnew99TvfHGG1a/fn07dOhQUHOapZ81rbSna5iZ9enTx3vqUSik1w/M/v6b1uyvDy9at27tXX799deH5O/F87Vt6rJ169ZZ9erV7YYbbrCPP/7Yrr76amvdunVIPtyKj4+3VatW2axZs6x3797eL2CSkpIsOTnZ7rrrLuvevftF759iGxmW2Vk7Dx48aK1atfL51jCYMpI39Y/D1atX2yWXXGI5c+a0qlWrpvthQrClHVD79+9v9erVs2LFiln16tXDIp/ZX4PmH3/8cdbyVatWWcOGDW3NmjUhSHVu33zzjd12221Ws2ZNa9GihcXHx9tvv/1md9xxh3Xr1i3k32CcOav3yJEjvZlGjBhhkZGR9sILL3hndv7zzz+tbt26NmbMmFDGNrP0v/3Zs2ePXXPNNSEvCM8cC9555x1v3rVr19q6det8viXYtGmT1apVKyTfbGdk3EpJSbH4+Hjr1KmT3XHHHWd9cxhM58u7Zs0aK1mypJUqVcpatGjhPbJo0KBB1rNnz6C/3zL6O2zAgAHeb4c6dOhgZcuWDfqYe+ZVKdL+Ppg0aZLlyJHDHnnkEe8HnUePHrXGjRv7FIrBcma7vv32286MW2e6/vrrvd9sX3fddVahQoWQfpN5Zj9Im2X79u22a9cunw84d+zYYXXq1LHPP/885FnPLApnz55trVq18o61J06csBtvvDEkv3vPd9WX9LIcOXLE2rZt6zPjezBl9Co1I0eOtGbNmllSUpJ16NDBypUrF/KxK72sqX1j48aNVr16dfN4PFazZk1v1mD2hzOfq0uXLnb//fd7758+fdpat25tgwcPvujnoNhGhmRm1s6kpCQ7evSoXX/99XbVVVeF5BdVZvKmvtFq165tzZs39+YNh0OFUj8B3Lt3rxUoUMCnPcMhX3qOHz9uHTt2tLZt24bF4e5pbdq0ySIiIuyaa67x+VZo27Zt3qyhKrjPNat32kLqf//7nxUoUMBatWplnTp1squvvtpq1qwZln3h2LFjduONN1rz5s3P+qMrmM41FqT3IVFiYqLt2rXLrr/+emvWrFnQ+29mZ0d+7bXXzOPx+HxQEEwZydujRw+76aabzvoW9sxv6sMha2o/feKJJ6x79+52yy23WOXKlb1/AAbrfXauq1Kkff63337brrjiCqtdu7Y1bdrUGjduHJKx4Fztmvb0ktGjR4f9uJXaD7t162avvPKKde7c2apUqeL92YdiDMtIP0iVkJBgmzdvtuuvv94aN24c9Lznypp2DF26dKlVrFjRrr/+erv77rutTZs2Vrt2bW8bB2u8zcxVX5KSkmz//v12/fXXW8OGDUPSZzOSN7XtXnvtNbv++uutY8eOVqlSpbAZu9Jr29QjTOvVq2dNmjQJ+d+2qRlfeOEFa9asmQ0dOtQ+/vhja9eundWtW/cf5aLYxgVlZtbO06dP24svvmht2rSxBg0ahOQXVWZnGT116pR16dLFSpYsGZSBKbO/UA4dOmS1atWyK664IuiDUWaynjhxwj788ENr3bp1SH6BXuh5Utdv3rz5nIevhurDgfPN6r1o0SJbtmyZ92c+f/58GzlypPXs2dOGDx/uXR7I91hm+8Hs2bMtJibG6tatG9I/VjMzFqSkpNi3335rXbp0CUnuzGY1++vblgceeCCs23bXrl3eb7TNfPtSsArtjGZNzTNkyBDzeDzWqFGjoP+xeqGrUqTNsWrVKnvvvfcsNjbWxowZE/TfD+dr1zOPGvjmm2+CPm6ZZX5Mb9eunXk8Hp8jyELxx/+F+sGZr2vlypXWtWtXq1+/ftDHrsxknTt3rvXs2dM6d+5sgwYNCnqfzcxVX06fPm2vvfaatW/f3mcsCOZ4m9mr1IwdO9Y8Ho81bNgw7MauM7OePn3abr/9ditevHhY/e29a9cue+ihh6xChQpWt25d69Klyz/+2VNs44I+/fRT69atm23dutXM/n4zXHPNNd5rFad9Ey1atMjnl2mwf1FlNu+pU6fsm2++CXjeu+++2z7++OOznj8jnnjiCe85hsFoz4vJevToUfvPf/5jDz30UFB/9v+kXcPFypUrbdy4cT6TRrVv396io6OtdOnS3m+vznU4WDj12aNHj9qLL75ojzzySMg/qc7sWLBz50777LPPvL9Qg5k7s1nP/HkEu+DOSN60QvnezGzbzpgxw+eb12D2g8OHD9vTTz9tK1asMDPzmSTzXIfnphXMfpCRdj1fnkC268WMXcnJyXbfffdZ69atQz52ZaQfpH1dBw8etCVLloRk7Mpsn02dTDVVMPtsZtt12bJlNnbs2JC068XknTt3rl177bVhO3ad+V7cuHFjWP3tnVqQnzhxwvbt22e7du3yPoZvthFQ/2TWzlB84xKOeffu3WtdunSxIkWK2Lx589J97vSc+UlcMM69udisZn9NMuGPgSmj/knWcJLRWb1TJ70Jxjfw/6Rt0/4xFcpDRcNxLDgX12ZHdilvZrKeeTpJKPpvRq5KcfDgwZCenmEWvn3gn4xd+/bt87ZrqOdGyejVSc7sB6E4Quti+2woxoaLzRqq99vF5g3XsSu9PhtOf3unt/6fvqcotnFemZ21c8qUKUHLlp5wzrt582a79957rVChQv/oTR8MF5M17WAZzF/2LrVrRoXLrN4u9YMzhfNYcCbXZkd2KW9ms06ePNm7XajHiXC+KkW494GLGbvSFifhdD75hfpB//79QxswjXDus2fKaLuGehxIdaG8Dz/8sHd9qIVTPwiHvxEptnHRwm3WzgsJVd60b9pNmzZZ7969M/SmT7v8008/tUWLFgU2qPkn62effeZM1mC1a2aEw6zeWbVtU7k0drmU1cytvC5kdeGqFGcKZbu69DssM1zqB2QNHJfyhjprOP0dEyEgk8xMklSgQAGlpKTopptu0vbt27Vx40blyJFDycnJIU7oK5R5zUwej0cpKSmSpEqVKmngwIHq0qWLunbtqvnz58vj8Xgznvk4SZowYYI6deqkyMjIgOX0Z9aOHTs6kzUY7ZpZqflS/fnnn7r//vuVnJysjh07Bvz5s3LbujR2uZRVciuvS1kjIyO978XBgwdr8+bNqlSpkn755RflzJlTSUlJIU74t1C3q0u/wzLLpX5A1sBxKW8os4bd3zH/uFxHluHarJ3hnjdtvqNHj/pc+mbHjh3Ws2fPdD9lS/up2oQJE6xQoUL24YcfBiwnWYMn3Gf1drVtw30sSMulrGZu5c3KWV25KoVZaNo1u4xdLvUDsmacS3nDPWs4jgUU23Bu1k4X8qbN9cwzz1iTJk3s0ksvtXbt2tkXX3xhSUlJtmfPHuvVq5cVKVLE5s+fb2a+hdSECROsQIECAf/FT9bAc2FWbxfb1oWxIJVLWc3cypvVs6Zy4aoUoWjX7DJ2pXKhH6Qi6/m5lNeFrOE6FlBsZ3OuzdrpWt6nnnrKihYtaq+++qq9//77dvXVV1uDBg3s9ddft5SUFNu2bZv16dPHPB6PLVu2zPu4V1991QoWLBjUgpCsgeHarN6utK1LY4FLWc3cypsdsrp2VYpQzeid1ccul/oBWS/MpbwuZTULv7GAYhvOzdoZznnTHo6ye/duq127tr377rve9QkJCda9e3erXbu2rVu3zszM1q1bZyNHjvTm2rp1q1122WX2wQcfBCwnWYMrnGf1drltw3ksOJNLWc3cypvVs4ZqduFwb9fsNna51A/ImjEu5Q3nrOE+FlBsZ2OuzdoZ7nnTFkQnT560Q4cOWaVKlWz69OlmZnbq1Cnv+goVKlhsbOxZ+0h90+/evTsgGcka2KxnCqfZMNPjatuG+1hwrucM96xnPm+4581uWbkqxd+y89jlUj8ga/pcyhvuWV0YCyi2s6nUN0HaTrpx40a75557zvsmSnv/tddes4iICFu8eHG2z5v2efr06WN33nmnHT582GrWrGk9evTwrks9hKZr16724IMPnnc/gULW4KDPBka4t6urWV3LS9bsm5Wxi37galbX8oZ7VlfGAortbCgcZ+rLKnm3b99u9erVswULFpiZ2dy5cy1v3rz2+OOPe7dJSUmxhg0b2tChQwOa5ULIGjj02cBwqV1dyupaXrKS1Yyxi6xuZXUtr0tZw30soNjOZtK+CcJppr6skPfFF1+0zp07W8+ePb2fosXHx9uECRMsd+7c1qpVK7vjjjusRYsWPpdACAWyBg59NjBcaleXsrqWl6xkNWPsIqtbWV3L61JWF8YCiu1sKtxm6ssKeQcNGmQFCxa0Zs2a+SxPTk62lStXWvfu3a1Xr14WGxvrfbMH8prJ50PWwKPPBoYL7epiVtfykjV7Z2XsIquLWV3L60JWF8YCiu1sIu3hHeE4U59reX/99Vfbs2ePmZnFxsba6tWrbf/+/fbss8+ax+OxF154wbvtud7Uwfp0jazBQZ8NjHBvV1ezupaXrNk3K2MX/cDVrK7lDfesro4FFNvZgAsz9aUVznlTUlJsw4YNVrRoUXv22WftgQceMI/HY2vWrDEzs/3799vTTz9t+fPnt1GjRvnkCcVkLGQNDvpsYIRzu7qc1cytvGTNnlkZu+gHLmc1cytvOGd1eSwwM4sQsjQzU0TEXz/m++67T/fee688Ho9y586tL7/8UpIUFRWl06dPS5IaNmyokydPnrWfyMhISVLp0qWzdV6Px6Nq1arpqaee0gsvvKCpU6dq/vz5qlWrlsxMxYsX1/3336/HHntMI0aM0JgxYyRJOXLkkMfj8WsWsoYm65nos4ER7u3qalbX8pI1+2Zl7KIfuJrVtbzhntXVsSAVxXYWl9rJfv/9d/3888+65557VKhQIb344ov64IMP9MQTT0iScubMKTPT1q1bVbRo0XPuJ7vnTU5OliSVK1dOOXPmVIECBbRixQrt3LnT+5wlSpRQnz59NGjQIMXGxmrGjBkByULW0GQ9E302MMK9XV3N6lpesmbfrIxd9ANXs7qWN9yzujoWeAX6q3OEngsz9YV73rSH15j9dYhNQkKCjR492sqUKWPDhw+3nTt3+myTkJBgkydPDnp7kjX46LOBEY7tmhWyupaXrNkrK2MXWbNCVtfyhmPWrDAWmHEYebawf/9+LVy4UJs2bVLOnDklSXnz5lWfPn30448/qmzZsoqKilLDhg21Zs0a5ciRw/spEnmllJQU7+E13377rb766iv9/vvvypUrlwYMGKB+/fpp8uTJevPNN7Vz505JUufOnfXbb7+pd+/eypEjh5KSkgKWj6yhR58NjHBr16yS1bW8ZM0+WRm7yJpVsrqWN9yyZpWxQBLfbGc1rs3UF+55006sEBsba6VLl7ZLLrnEWrRoYc8//7x33XPPPWeXX365dezY0Zo2bWrR0dHeTwaDhazBQZ8NjHBvV1ezmrmVl6zZNytjF1ldzWrmVt5wz+rqWHAuFNtZhGsz9bmQN+3zLFu2zOrVq2fLly+3pUuX2sMPP2z16tWzJ5980rvN5MmTbcCAAfbggw96B6FgDZxkDU5u+mxgMod7u7qY1bW8ZM3eWRm7yOpiVtfyupDVxbHgQii2s5ixY8daoUKFLHfu3PbVV1+Z2d8dd9++ffbMM89YwYIFbfTo0SFM+TcX8s6YMcO6devmc5mDPXv22ODBg+3KK6+0YcOGeZen/UQtFG92sgYefTYwXGjXVC5lNXMrL1kDw5WsjF2BRdbAcSmvC1ldHAvOJUeoD2OHfyQnJysyMtI7U1+uXLm0YsUKVa1aVWXLlpX090x9ERERio2NValSpdS1a1fynsehQ4c0c+ZMffvtt2revLl3ealSpfTwww9Lkr744gsdP35cL730kvc8F+mvSw6Q1f2sqeizgeFKu7qW1bW8ZCUrYxdZXcvqWl5Xsro2FlxQqKt9/DOuzdQX7nlTP9lLexjLxo0b7Z577rEyZcrYq6++6rP93r177cEHH7SePXuG5HAgsgYefTYwwr1d03Ipq5lbeckaGC5kZewKPLIGjkt5wz2rq2NBRlFsOyztm2fRokU2f/5827hxo3fZyJEjrUyZMvb000/bjh07zMysU6dO3nMzzIJ7uEW4502bLy4uzg4dOmTx8fFmZrZp0ybr2bOnNWnSxCZOnOjzuAMHDqQ7UAQSWYODPhsY4d6urmZ1LS9Zs29Wxi6yuprVtbzhntXVsSAzKLYd5dpMfeGeN22+ESNGWP369a1atWpWv359W7RokZmZbd682Xr27GlNmza1119//bz7IKt7Wc/3vPRZ/wn3dnU1q2t5yZp9szJ2kdXVrK7lDfesro4FmUWx7SDXZupzKe+IESOsaNGi9tZbb9nkyZOtW7duFhUVZdOmTTMzsw0bNtg999xjlSpVso8//jgomcgafPTZwHCpXV3K6lpespLVjLGLrG5ldS2vS1ldGgsuBsW2w1ybqS+c8yYlJdmhQ4esYcOGNmnSJJ91AwcOtFy5ctnatWvNzGzt2rX23//+95zXHgw0sgYPfTYwwrldz+RSVjO38pI1MMI9K2NXcJA1cFzKG85ZXR4LMoNi21EHDx60m2++2YoVK2adO3f2WZf6JmrQoIHPmyuUwjHv4cOHbdeuXT45ihcvbrNmzTIz30GnefPm9sADD5w1yUSw3vRkDT76bGCEY7uei0tZzdzKS9bACNesjF3BRdbAcSlvOGbNCmNBZkWEejZ0ZIyZ+fxbpEgR/fe//1XHjh21YsUKvfbaa95tS5Uqpf79+6thw4Y6dOiQ9zHk/dvs2bPVu3dv3XbbbZozZ443x5VXXqlJkybp1KlTypkzp5KSkmRmKl68uFJSUhQR4fuWiYyMJKujWc9Enw2McG9XV7O6lpes2TcrYxdZXc3qWt5wz+rqWPCP+b9+h7+5NlNfuOedPHmyFS1a1F599VVbsmSJz7r33nvPrrrqKuvbt693WXJysl199dU2ePDggGU6F7IGB302MMK9XV3N6lpesmbfrIxdZHU1q2t5wz2rq2OBP1BshznXZuoL97xfffWVFS9e3N59912f5amD1MmTJ+1///uf1atXz6pWrWo9evSwRo0aWfXq1YN+rg1Zg4M+Gxjh3q6uZnUtL1mzb1bGLrK6mtW1vOGe1dWxwF8oth3h2kx94ZY3dRAZOHCg3Xzzzd5P+9JKfdOfOnXKli5dag899JDdc889NnjwYO+bPRjniZA1NOizgRFu7ZpVspq5lZes2ScrYxdZs0pWM7fyhlvWrDIW/FMU22HOtZn6wjlvQkKC1alTxx566KF016cOCvv37093fTA/XSNr8NBnAyOc2/VMLmU1cysvWQMj3LMydgUHWQPHpbzhnNXlscBfKLbDkGsz9bmUt3Xr1talSxdLSEhId/2JEyesffv2NmfOnKDkOR+yBg59NjBcaleXspq5lZesZDVj7CKrW1nN3MrrUlaXxoJAYDbyMOPaTH2u5a1SpYoWL16s9evXe5dZmhkYd+7cqdy5c+vyyy8PSp7zIWtg0GcDw6V2dSmra3nJStZUjF1kdSWra3ldyiq5MxYECsV2GJkyZYr69OmjNm3a6KWXXlLnzp2963r27KkjR47o0UcflSTlyJFDZqaDBw+qSJEi5L2A1Df1E088oSJFiqhnz55av369Tp8+LY/Ho5SUFMXHx+uxxx5TcnKyateuHfSMZA08+mxguNSuLmV1LS9ZySoxdpHVrayu5XUpq0tjQUAF7ktzZIZrM/W5ljdtvs8//9wqVKhg5cuXtxEjRtjXX39tL7/8srVp08Zq1qzpPfTmzMNtyOp2VvpsYLjUri5ldS0vWcl6JsYusoZ7VtfyupT1zHzhPBYEmscsBFeIh5eZyePxaNCgQdq6daveeust5c2b12eb1EM/EhIStHr1ar311ls6deqUSpQooWeeeUY5cuRQcnJyUA4HcS1vepKSkrRu3ToNHTpUS5cu1cGDB9WoUSPVrFlTEyZMUI4cOZSUlKQcOXKEJB9Z/Ys+GxgutatLWV3LS1ayng9jF1nDMatreV3Kei7hOBYETfDre5zJtZn6XMt7Prt377YNGzb4XI4gXC8xQNaLR58NDJfa1aWsZm7lJWtguJQ1Ixi7Lg5ZA8elvC5lvZBwGguCgXO2w0CuXLlUtGhRxcXFKTEx8az1Ho9HJ0+eVI8ePfTxxx+ftT7YnwKFa1474yCNM++nlZycLEkqXbq0qlWr5v2E0MyC9onq+e6nRdZ/jj4bGOHarulxKavkVl6yBkY4Z2Xsoh+kx6Wsklt5wzWr62NBMFBshwnXZuoLx7wej0eStHnzZu/9c73pU9/Uqevtr8vgefcRaGQNPvpsYIRju56LS1klt/KSNTDCNStjV3CRNXBcyhuOWbPCWBBw/vuSHBcj9bCPnTt3WtWqVa1OnTq2bt06n4kCjh8/bp07d7aOHTuGfOKAcMyb9jmmT59urVq1ss8///yszGdKuzzt9QgDiazBR58NjHBs16yQ1cytvGTNXlkZu8iaFbKauZU3HLNmhbEgWJggLUykpKRo7ty5+te//iUzU8+ePdWsWTOtW7dOn376qeLi4rRy5UrlzJkz3WvlZde8afc9f/58zZ49W++//77q1aunIUOGqG3btpJ01idnae+PHj1aM2bM0Pfffx/QQ4LIGlr02cAIl3bNalldy0vWrJ+VsYt+kNWyupY3XLJmtbEg4IJT0yMjTp8+batXr7YbbrjBihUrZh6Pxxo3bmy9e/f2TmwQThMchFPeRx991MqUKWMjRoywgQMHWvHixS0mJsa++OIL7zapn6al/VRtwoQJVqRIEXvnnXeCkpOsoUWfDYxwatcLcSmrmVt5yRoY4ZaVsSs0yBo4LuUNp6xZaSwIJIrtMOXaTH2hzLtq1SorXbq0ff31195lP/74ozVu3NhatWplX331lXd52gFowoQJVqBAAfvwww+DkpOs4YU+GxgujV0uZTVzKy9ZAyPUWRm7wgNZA8elvPwd4waK7SA487yFc53HYHbuN8n5HuNvruXduHGjlSpVyvvGTn3uJUuWWO7cua1t27b25Zdf+jxm4sSJIXmzkzUw6LOB4VK7upQ1vecK57xkDQyXsqZi7PI/sgaOS3ldymrmzlgQDpiNPAhcm6kvnPOml8Pj8cjj8WjDhg2S/rq0gJnpqquuUt26dbV79269+eab+u233yRJ06ZN0wMPPKA333xTN998c0BykjVwWdNDnw2McG7XM7mUNTWf5EZesgZGuGdl7KIfnMmlrKn5JDfyhnNWl8eCsPDPanWcj2sz9YV73rT59u7daydOnPB+uvfss89ajhw5bPbs2d5tjh8/bnfffbe98cYbVrhwYRs7dqyZmX300Uf22WefBSwnWYOHPhv43OHYrmm5lNXMrbxkDQwXsjJ2BR5ZA8elvOGe1dWxIJxQbAdI2s45b948e+CBB6xgwYLWqlUrmz9/vnfd+Q4bGTVqlDVu3DgoEx24lPepp56yGjVq2FVXXWUDBgywEydOmJlZ//79zePx2EMPPWRPPPGEtWzZ0urXr29mZh06dLDbbrvNzCyol28ga+DQZwPDpXZ1KatreclKVjPGLrK6ldW1vC5ldWksCDcU2wHm2kx94Zg37Rv07bfftmLFitmUKVPsoYcessaNG1tMTIydPHnSzMxef/11i4mJsRYtWtjtt99uCQkJZmbWunVrGzp0qN+zkTU4Wc+HPhsY4diuWSGra3nJmr2yMnbRD7JCVtfyhmPWrDAWhAuK7QBybaa+cMyb9s3+6aef2vPPP2/vvfeemf01QcSHH35o9evXt7Zt23pnYzx69Kj3MSdPnrTHH3/cSpYsaRs3bvR7PrIGPuv50GcDIxzbNStkdS0vWbNXVsYu+kFWyOpa3nDMmhXGgnBCsR1Ars3UF05527dvb7/88ov3/ooVK6xKlSpWoEABn3NDEhMT7aOPPrIGDRrYdddd53P5g82bN1v//v2tTJkytnLlSr/mI2vgs2YEfTYwwqlds1JW1/KSNXtkZeyiH2SlrK7lDaesWWksCCcU236S3gQGv/32m5UuXdpeeeUVM/vrE6nU7a666iqrXr263Xbbbd5PfaZOnWoej8c++uijbJ13+/bt9thjj3kPQzEzO3TokL3yyitWrlw5u/766322T0xMtFmzZtlll11mAwYM8C6Pj4+3FStW2O+//+7XfGQNfNb00GcDI5zb1eWsruUla/bMythFP3A5q2t5wzmry2NBuKPY9gPXZupzKe9LL71kCxcuNDOzY8eO2auvvmq1atWynj17+myXmJhoixYtOue1B4OBrIFDnw0Ml9rVpayu5SUrWc0Yu8jqVlbX8rqU1aWxwAUU237k2kx94Zg37T5Pnjxp1113nRUqVMh++OEHMzM7cuSIjRs3zurWrWu9evVKdx/BetOTNfjos4ERju2aFbK6lpes2SsrYxf9ICtkdS1vOGbNCmNBOIsI9XW+XZaSkuL9/zvvvKNx48YpNjZW9erV05IlS3TTTTfp1KlTGjNmjCZNmqQtW7bou+++U3R0tH788UdJ0qlTp1SlShVJUkREYH8c4Z5369atOn36tCTpxRdflJlp7NixuuGGG9SpUyf98MMPKliwoO666y7de++9WrNmjTp37nzWfiIjI/2ai6zBy3om+mxghHu7uprVtbxkzb5ZGbvoB65mdS1vuGd1dSxwSigrfZe5NlNfuOf98ccfrXz58vbhhx/aww8/bB6PxzZt2mRmZhs2bLBu3bpZsWLF7Pvvvzezvz5le+6556xnz55B/wSVrMFBnw2McG9XV7O6lpes2TcrYxf9wNWsruUN96yujgWuodjOJNdm6nMpb7du3ax48eKWP39+W7p0qc+69evXW7du3axEiRLew1qOHz/unUQi2G96sgYOfTYwXGpXl7K6lpesZDVj7CKrW1ldy+tSVpfGAldRbGeCazP1uZI39bqBEyZMsHz58ln58uVtxowZduzYMZ/t1q9fb3feead5PB5bs2aNd3l6szsGClkDiz4bGK60q2tZXctLVrIydtEPXMvqWl5Xsro2FriMYvsiuTZTXzjmPfMTsU2bNtnvv/9ud9xxh1WrVs2mTp1qx48f99lm+/bt9vjjjwe9PckafPTZwAjHdj0Xl7KauZWXrIERrlkZu4KLrIHjUt5wzJoVxgLXUGxnkGsz9YV73rT5Vq5caT///LOtWLHCu+z222+3atWq2dtvv+0zU2Pq/83+/lQu0MgaHPTZwAj3dnU1q5lbecmafbMydpHV1axmbuUN96yujgWuo9jOgC1bttipU6fMzOx///ufnThxwn777Te76667zpo4YNy4cVa/fn3r1KkTec8h7aEnTzzxhNWsWdMqV65sl156qT300EPedXfccYdVr17d+vbta23btrWCBQsG/U1O1uCgzwZGuLerq1ldy0vW7JuVsYusrmZ1LW+4Z3V1LMgKKLYvwLWZ+lzK+9xzz1nRokXthx9+sPj4eHvsscfM4/HYsmXLvNvExsbabbfdZrfeeqslJiaaWWiu5UfWwKHPBoZL7epSVtfykpWsZoxdZHUrq2t5Xcrq0liQVVBsZ4BrM/W5kDcpKcm6du1q06ZNMzOzWbNmWaFChWzChAneTKlCffgKWQOPPhsYLrSri1ldy0vW7J2VsYusLmZ1La8LWV0cC7ICiu3zcG2mvnDOe+a+//zzTytbtqzNmTPHFi5caPnz57fXXnvNzP6aKGLo0KH21VdfnXcfZHUra3ros4ERzu3qclbX8pI1e2Zl7CKry1ldyxvOWV0eC7ISiu10uDZTn0t5075pH3vsMWvfvr3lzZvXXn/9de/yuLg4a9eunU2cODGo2c5E1sChzwaGS+3qUlYzt/KSNTBcymrG2BUoZA0cl/K6lNWlsSArotg+g2sz9YV73rT5Xn/9datXr573/I/p06db2bJlLSYmxrZv325mZvv27bMOHTpYs2bNQjpwkjVw6LOBEe7t6mpW1/KSNftmZewiq6tZXcsb7lldHQuyKortNFybqS/c86Z9s3/++ef23//+1zwej3Xq1MmbfezYsVa1alWrWbOmNW3a1Bo1auQzKITicg1kDRz6bGCEe7u6mtW1vGTNvlkZu8jqalbX8oZ7VlfHgqyMYjsdrs3UF+55Bw4caBUqVLBhw4ZZly5drFChQtaiRQvv+m+++cZef/11e+KJJ+ztt9/25grFgE/W4KDPBka4t6urWV3LS9bsm5Wxi6yuZnUtb7hndXUsyIoots/g2kx94Z53yZIlVrRoUfv666+9z/v111/bZZddZtdcc805Z2AMxcBJ1uCgzwZGuLdrWi5lNXMrL1kDw4WsjF2BR9bAcSlvuGd1dSzIqrJ9se3aTH3hnvfMN/D3339vRYoUsX379nmXnT592mbPnm0ej8duuukm7/JQT25B1sCgzwZGuLerq1ldy0vW7JuVsYusrmZ1LW+4Z3V1LMgusn2xncq1mfrCPe/gwYPtmWeesb1791rp0qVt/PjxPut37dpllStXthw5cljbtm2Dni8tsgYHfTYwwr1d03Ipq5lbeckaGC5kZewKPLIGjkt5wz2rq2NBVhehbColJcX7/8mTJ6tBgwY6ffq0JOnKK6/U2rVr1aJFC7Vt21aStH//ft1zzz36888/1bt3b/KeJ99nn32mDz/8UK1bt1a+fPnUoUMHzZ49Wx988IF3m9y5c6tRo0aaOXOmfvvtN02cODHgGckaXPTZwAj3dnU1q2t5yZp9szJ2kdXVrK7lDfesro4F2U2OUAcIhZSUFEVE/PU5wxdffKH9+/dr1apVuvXWWzV79mx17dpVf/zxh8aNG6cbbrhBBQoUUFJSkpKSkrR06VJFRkYqOTlZkZGR5P3/UvPNnTtXn376qe666y41bdpUktS/f38NHTpUL730khYvXqwmTZpo0qRJioiI0DXXXKMCBQpo586dActG1uCjzwaGC+3qYlbX8pI1e2dl7CKri1ldy+tCVhfHguzIY2YW6hChMmjQIH300Ue66667tHbtWn3zzTeqVauWvvvuO0nSwoULtWXLFm3fvl3VqlVTt27dFBkZqaSkJOXIEfzPKcI9744dO9ShQwdt3bpVPXr00GuvveZd99tvv+mjjz7StGnTlD9/fhUvXlwff/yxoqKi1KZNG7Vv317//ve/ZWbyeDxkdTTrmeizgRHu7epqVtfykjX7ZmXsIqurWV3LG+5ZXR0LspXQHcEeWq7N1OdK3h9//NGaNWtm1apVs88///ys9adPn7Zjx4557w8aNMhKlSplW7ZsCWZMMyNroNFnA8OVdjVzK6uZW3nJGhguZWXsChyyBo5LeV3J6tpYkN1km2LbtZn6wj1v2nxnzrC4ePFia9asmXXq1MlnNsa0lzxYvny59ezZ0y699FJbuXIlWR3Meib6bGCEe7um5VJWM7fykjUwXMjK2BV4ZA0cl/KGe1ZXx4LsLtsU26lcm6kvHPOmfbNPmjTJ+vXrZ/fee699+umndurUKTMz++abb6xZs2bWuXNn7yeCZ3rvvfcC/qkaWYOPPhsY4diu5+JSVjO38pI1MMI1K2NXcJE1cFzKG45Zs8JYkF1l+WI7bef89NNPrVKlSvbDDz/YsWPH7N5777U2bdrYzJkzvdscOHDA7rzzTps1a5Zddtll3gvUk/dsgwYNsuLFi1v//v0tJibGGjdubEOHDrWTJ0+amdnChQvt6quvthYtWtiKFSvSfY1kdTvrmc9Ln/Ufl9rVpayu5SUrWc0Yu8jqVlbX8rqU1aWxAH/J8sV2qi+//NLuu+8+e+qpp7zLfv31V+vUqZM1btzY/vWvf9l7771nLVu2tNatW9vBgwetZs2a9sQTT5A3HVOmTLEKFSrYTz/9ZGZms2bNssjISKtRo4Y9+uij3jf9l19+aQ888EBI3+RkDQ76bGCEe7u6mtW1vGTNvlkZu8jqalbX8oZ7VlfHguwuWxTbv//+u9WoUcPy5MljDzzwgM+6jRs32n//+1+rUqWK1atXz2JiYryHY1x77bX24osvmtnZ50Zkt7yp+0v9d9SoUTZixAgz++vNXqhQIRs1apT17dvXihUrZo8//ridOHHCZx/BetOTNfjos4ERju2aFbK6lpes2SsrYxf9ICtkdS1vOGbNCmMBskmxbebeTH3hlDcuLs77/48++sjMzI4cOWJ79uyxnTt3Wq1atbwDzebNm61EiRJWtmxZGz16tJkFd2Ana+jQZwMjnNr1QlzKauZWXrIGRrhlZeyiH1yIS1nN3MobTlmz0liQ3WW5Ytu1mfrCPe+XX35pbdq0sV9//dUGDBhgHo/Hdu3a5V3/9ddfW8WKFW3Tpk1mZrZs2TK75ZZbbMKECUH/NI2swUGfDYxwb1dXs7qWl6zZNytjF/3A1ayu5Q33rK6OBUhfliq2XZupz4W8GzdutMqVK1vFihWtcOHCtmbNGjP7+xIHCxcutGrVqtno0aNty5YtdsMNN1ifPn28g1cwL9tA1sCjzwaGC+3qYlbX8pI1e2dl7KIfuJjVtbwuZHVxLMC5ZaliO5VrM/WFa97UN+u///1vy5kzpzVv3tx+/vlnn22OHDliPXv2tPLly1upUqWsYcOGlpiYaGbBPYSFrMFFnw2McG1X17O6lpes2S8rYxdZXc/qWt5wzer6WICzZbli27WZ+lzIO2vWLPviiy+sVq1a1r59e/vuu+981h85csR+/fVXmzt3rneQSHu4TTCRNfDos4HhQru6mNXMrbxkzd5ZGbvI6mJWM7fyupDVxbEA6XO+2HZtpj7X8qa1fv16q1GjhrVv396+//577/IpU6b4bBcOh6+Q1X/os4HhUru6lNW1vGQla3oYu8gazlldy+tS1jOF81iAjHG62HZtpj7X8pr9PbikTsKwceNGq1WrlsXExNjYsWPtxhtvtEKFCoX8E1UzsgYCfTYwXGpXl7K6lpesZE2LsYusLmR1La9LWVO5MBYg45wttl2bqc+1vGZ/DzAffPCBFS9e3H799VczM/vtt98sJibGmjVrZm3btvWeJxLKNz1Z/Y8+GxgutatLWV3LS1aypsXYRVYXsrqW16WsqVwYC5A5zhbbrs3UF855z/dGnT17tuXKlcvGjRvns+2xY8ds37593nzBOk+ErMFDnw2McG5Xl7O6lpes2TMrYxdZXc7qWt5wzuryWIDMcbLYdm2mvnDOm3bfb731lj3zzDM2ZcoU279/v5mZPfbYYzZp0qRzPsYs+OfcmJE10OizgRHO7epyVtfykjV7ZmXsIqvLWV3LG85ZXR4LkHlOFtupXJupL9zypn3jDhw40IoXL25XXnml1ahRw6655hrbvXt3wJ47s8gaGvTZwAi3dj0fl7KauZWXrIERjlkZu4KPrIHjUt5wy5pVxgJknNPFdirXZuoLt7xbt261u+66y3755RdLSkqyzz//3Fq3bm316tWz7du3BzXLhZA1NOizgRFu7Xo+LmU1cysvWQMjHLMydgUfWQPHpbzhljWrjAW4MGeLbddm6gvXvO+8845dccUV1qpVKzty5Ih3+YIFC+zaa6+1evXq2e+//+7zGkKFrMFFnw2McG3X9LiU1cytvGQNjHDOytgVPGQNHJfyhmtW18cCZE6EHGRmioiI0IcffqimTZtq7dq1qlq1qj788ENJ0vTp03Xq1Cnt379fERERSklJIe85nDp1SgULFtT69esVGRnpXd66dWs9/vjjKlasmJo2bap9+/YpIiK03YWswUOfDYxwbleXs7qWl6zZMytjF1ldzupa3nDO6vJYgIsQigo/o1ybqS/c86aX7/Tp0zZjxgyrVq2aXXfddXbw4EGf9V988YX1798/6IeykDU46LOBEe7tmpZLWdNmSE+45SVrYLiQlbEr8MgaOC7lDfesro4F8B+PmVmoC/70mJk8Ho8k6e2339bvv/+u0qVL68Ybb1Tx4sU1ePBgVaxYUX369En3MZKUkpIStE+Ewj1v2n2vWLHCe79hw4YyM33wwQcaM2aMChcurHfeeUeFCxc+ax/Jyck+n8AFClmDgz4bGOHerq5mdS0vWbNvVsYu+oGrWV3LG+5ZXR0L4GeBreUvjmsz9YV73rT5Bg0aZGXLlrXLLrvMoqKi7J577rEtW7aYmdn06dOtadOmduONN9qBAwfImoWynok+Gxjh3q5puZTVzK28ZA0MF7IydgUeWQPHpbzhntXVsQD+F5bFdirXZuoL97xjx461YsWK2Q8//GAbN260r776yqKjo61Lly62f/9+S05OtnfffdeqVKlijz76aMhykjV46LOBEe7tmpZLWc3cykvWwHAhK2NX4JE1cFzKG+5ZXR0L4D9hW2y7NlOfC3m7d+9u999/v5n9/YnbqlWr7JJLLrEnnnjCzP46j2T+/PkhH0TJGnj02cBwoV1TuZTVzK28ZA0MV7IydgUWWQPHpbwuZHVxLIB/he0Ud67N1Bduee2MU/FPnz6t3bt369SpU971iYmJqlu3rp566inNnDlTBw8eVI4cOdS2bVtFRkYqOTk54DnJGjr02cAIt3Y9H5eySm7lJWtghGNWxq7gI2vguJQ33LJmlbEAfhaaGt+XazP1hXvetPm2bNli+/btMzOzadOmWb58+ezrr782s78/YRs3bpw1adLEEhISAp6NrKFBnw2McG/XtFzKauZWXrIGhgtZGbsCj6yB41LecM/q6liAwAt5sZ22cy5fvtyWLl1qy5cvN7O/OuSMGTOsSZMm1qFDBzt06FC6+wjmG96lvEOGDLEaNWpY0aJFbeDAgfbRRx/Zv/71L6tatap9+eWXlpSUZEeOHLF27drZLbfc4jOZQ7CRNXDos4HhUru6lNXMrbxkJasZYxdZ3cpq5lZel7K6NBYgOEJabLs2U1+45007GM2cOdPKlCljs2fPthEjRliTJk2sa9euNnr0aHv00UctMjLSKleubNWqVbM6depYYmLiWa+RrO5lPRN9NjDCvV1dzWrmVl6yZt+sjF1kdTWrmVt5wz2rq2MBgifk32ybuTdTX7jn/fbbb+3hhx+2KVOmeJd9/PHH1qZNG7v11ltt/vz5tnr1aps4caK9++673k/7Tp8+TdYskvVM9NnACPd2dTWra3nJmn2zMnaR1dWsruUN96yujgUIvLAotl2bqS+c8+7du9cqVqxoBQoUsNGjR/us++STT6x169bWuXNnW7Zsmc+6ULQrWYOHPhsY4dyuZ3Ipq5lbeckaGOGelbErOMgaOC7lDeesLo8FCLygF9tnHiqRmJho1157rfXo0cPM/jocI3WygJdeeskqV6581uEgweycruU1M/vll1+sSpUq1rZtW1uzZo3Pus8//9xq1qxpgwcPDmqmcyGr/9FnA8OldnUpq5lbeclK1rQYu8jqQlYzt/K6lDWVC2MBQiOoc/anpKTI4/FIkrZu3ar9+/crZ86c6t69uz788EMtWLBAERERypkzpyQpKipKxYoV0yWXXOKzn7TT+5P3bLVr19bMmTN14MABjR07VuvWrfOu69ChgyZOnKj//Oc/Qc10LmT1L/psYLjUri5ldS0vWcl6JsYusoZ7VtfyupQ1rXAfCxA6QS22U69x9/jjj6tjx46qXr26Bg0apPz58+uee+5R3759NXfuXKWkpOjo0aP67LPPVKZMGe8bKthcy5tWnTp1NGXKFP388896+eWXtX79eu+6pk2bhtW1/MjqP/TZwHCpXV3K6lpespI1PYxdZA3nrK7ldSnrmcJ5LEDoeMzOuAJ7AKSkpHjfPB988IEeeeQRjRs3TmvWrNHcuXN12WWX6aqrrtLu3bs1evRoVahQQZGRkYqKitKKFSuUM2dOmZn3ky7yZtyqVat0//336/LLL9cLL7yg8uXLhzrSOZH14tFnA8OldnUpq2t5yUrWjGDsIms4ZXUtr0tZLyScxgKEXlCK7VTfffedPvroI9WpU0f33HOPJOmTTz7R2LFjVbhwYfXp00clSpTQsmXLlD9/fnXt2lWRkZFKSkpSjhw5ghXT2bznsnz5ck2YMEGTJ0/2DmThiqz/DH02MFxqV5eyupaXrGS9EMYusoZbVtfyupT1fMJtLEAIBevkcNdm6nMt74WkTjaR9nqA4YqsF4c+GxgutatLWc3cykvWwHApa0YxdmUeWQPHpbwuZc2IcBkLEFpBnY3ctZn6XMt7IWfO7hjOyHpx6LOB4VK7upTVzK28ZA0Ml7JmFGNX5pE1cFzK61LWjAiXsQChE/RLf61evdquvPJK69Onj61du9Zn3Q8//BB2n065lhegzwaGS+3qUlYzt/KSNTBcyuoal9qWrIHjUl6XsgIXEvRi28xs5cqVVq9ePevTp4+tW7furPXh9iZyLS9Anw0Ml9rVpaxmbuUla2C4lNU1LrUtWQPHpbwuZQXOJyTFttlfb6KGDRvaLbfcYlu3bg1VjAxzLS9Anw0Ml9rVpaxmbuUla2C4lNU1LrUtWQPHpbwuZQXOJWTT41155ZUaN26cLrnkEl1++eWhipFhruUF6LOB4VK7upRVcisvWQPDpayucaltyRo4LuV1KStwLkG99Fd67P9fEy/t9fXCmWt5AfpsYLjUri5lldzKS9bAcCmra1xqW7IGjkt5XcoKnCnkxbaksLkIfUa5lhegzwaGS+3qUlbJrbxkDQyXsrrGpbYla+C4lNelrEBaYVFsAwAAAACQlXAsBgAAAAAAfkaxDQAAAACAn1FsAwAAAADgZxTbAAAAAAD4GcU2AAAAAAB+RrENAEAWsmjRInk8Hh05cuSc20ydOlWFChUKWiYAALIjim0AAMJEXFyc+vfvr0qVKil37twqWbKkmjVrptdee00nTpzI0D6aNm2qvXv3qmDBggFOCwAAzidHqAMAAABp69atatasmQoVKqRnn31WtWrVUlRUlH799VdNmjRJZcqUUceOHS+4n1y5cik6OjoIiQEAwPnwzTYAAGHgoYceUo4cOfTTTz/ptttu0xVXXKEKFSqoU6dO+vzzz3XjjTdq+/bt8ng8Wr16tfdxR44ckcfj0aJFiySlfxj51KlTddlllylv3ry66aabdPDgwbOe/+OPP1a9evWUO3duVahQQSNGjFBSUlKAXzUAAFkXxTYAACF28OBBzZ8/X3379lW+fPnS3cbj8VzUvpctW6bevXurX79+Wr16tVq1aqX//Oc/PtssXrxY3bt3V//+/bV+/XpNnDhRU6dO1X//+9+Lek4AAECxDQBAyG3evFlmpqpVq/osL1asmPLnz6/8+fPrscceu6h9v/zyy2rXrp0GDRqkKlWq6OGHH1ZMTIzPNiNGjNDgwYPVo0cPVahQQW3bttUzzzyjiRMnXvRrAgAgu6PYBgAgTC1fvlyrV69WjRo1lJCQcFH72LBhgxo3buyzrEmTJj73f/nlFz399NPewj5//vzq06eP9u7dm+GJ2QAAgC8mSAMAIMQqVaokj8ej3377zWd5hQoVJEl58uSRJEVE/PUZuZl5tzl9+vQ/fv7jx49rxIgR6tKly1nrcufO/Y/3DwBAdsQ32wAAhFjRokXVtm1bjRs3TvHx8efcrnjx4pL+Xzt3jNJIFMBx+L/pvERCMI2NgSTCwLSCjWDjOYKFTUAkkCZFQERIE/AUKTQnUCwkkEJPIHiGwG61i5Jl3WVnQZbvg1fM4zEw5e895iUvLy8/5t5elvYzOzs7ub+/fzd3d3f37rnT6eT5+TmtVmtjfA98AODPONkGgE9gOp2mLMv0er0Mh8Ps7u6mVqvl4eEhT09P6Xa72draSlEUGY/HaTabeX19zdnZ2S/f2+/3U5ZlJpNJjo6Ocnt7m5ubm3drzs/Pc3h4mHq9nuPj49RqtSyXy6xWq43L1ACA32O7GgA+ge3t7Tw+PmZ/fz+DwSDtdju9Xi9XV1c5PT3NaDRKklxfX2e9Xqfb7ebk5OTDGC6KIrPZLJeXl2m321ksFhuBfnBwkPl8nsVikb29vRRFkYuLizQajX/2vQDwv/vy9e2PXwAAAMBfc7INAAAAFRPbAAAAUDGxDQAAABUT2wAAAFAxsQ0AAAAVE9sAAABQMbENAAAAFRPbAAAAUDGxDQAAABUT2wAAAFAxsQ0AAAAVE9sAAABQsW8WRc01RnAl5gAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# sum the number of -1 per row in pd.DataFrame(outlier_scores) and make into barplit\n",
    "#subset to only first 20 guides\n",
    "outlier_counts = pd.DataFrame(outlier_scores).apply(lambda x: (x == -1).sum(), axis=1)\n",
    "outlier_counts = outlier_counts.head(20)\n",
    "\n",
    "outlier_counts = outlier_counts.sort_values(ascending=False)\n",
    "plt.figure(figsize=(10, 6))\n",
    "sns.barplot(x=outlier_counts.index, y=outlier_counts.values, palette=\"viridis\")\n",
    "plt.xticks(rotation=45)\n",
    "plt.xlabel(\"Guide\")\n",
    "plt.ylabel(\"Number of Outliers\")\n",
    "plt.title(\"Outlier Counts by Guide\")\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "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>index</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   index  count\n",
       "0      1     18\n",
       "1      2      2"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1000x600 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#create a count plot of the number of outlier \n",
    "plt.figure(figsize=(10, 6))\n",
    "county=pd.DataFrame(outlier_counts.value_counts()).reset_index()\n",
    "county.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:matplotlib.category:Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
      "INFO:matplotlib.category:Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "sns.barplot(county, x=\"index\",y=\"count\", palette=\"viridis\")\n",
    "#add line at y=3\n",
    "plt.xticks(rotation=45)\n",
    "plt.xlabel(\"Count of Outliers\")\n",
    "plt.ylabel(\"Number of Guides\")\n",
    "plt.title(\"Number of times guide assigned as an outlier\")\n",
    "#make y axis to lavel every two\n",
    "plt.yticks(np.arange(0, county['count'].max()+1, 2))\n",
    "#save as pdf\n",
    "plt.tight_layout()\n",
    "plt.savefig(loc+\"figures/number_of_outliers_per_guide.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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>IsolationForest</th>\n",
       "      <th>EllipticEnvelope</th>\n",
       "      <th>LocalOutlierFactor</th>\n",
       "      <th>OneClassSVM</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>non-targeting_00084</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_00202</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_00303</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_00362</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_00373</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_00715</th>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_00759</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_00790</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_00850</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_00867</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_00949</th>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_01357</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_01500</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_01651</th>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_01660</th>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_01777</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_01815</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_01967</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_02135</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_02295</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_02541</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_02771</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_02790</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_02839</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_02847</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_02969</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_03155</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_03351</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>non-targeting_03447</th>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     IsolationForest  EllipticEnvelope  LocalOutlierFactor  \\\n",
       "non-targeting_00084               -1                 1                  -1   \n",
       "non-targeting_00202               -1                 1                  -1   \n",
       "non-targeting_00303               -1                 1                  -1   \n",
       "non-targeting_00362               -1                 1                   1   \n",
       "non-targeting_00373               -1                 1                   1   \n",
       "non-targeting_00715               -1                -1                  -1   \n",
       "non-targeting_00759               -1                 1                  -1   \n",
       "non-targeting_00790               -1                 1                  -1   \n",
       "non-targeting_00850               -1                 1                  -1   \n",
       "non-targeting_00867               -1                 1                  -1   \n",
       "non-targeting_00949               -1                -1                   1   \n",
       "non-targeting_01357               -1                 1                   1   \n",
       "non-targeting_01500               -1                 1                  -1   \n",
       "non-targeting_01651               -1                -1                  -1   \n",
       "non-targeting_01660               -1                -1                  -1   \n",
       "non-targeting_01777               -1                 1                  -1   \n",
       "non-targeting_01815               -1                 1                   1   \n",
       "non-targeting_01967               -1                 1                   1   \n",
       "non-targeting_02135               -1                 1                   1   \n",
       "non-targeting_02295               -1                 1                  -1   \n",
       "non-targeting_02541               -1                 1                  -1   \n",
       "non-targeting_02771               -1                 1                   1   \n",
       "non-targeting_02790               -1                 1                  -1   \n",
       "non-targeting_02839               -1                 1                  -1   \n",
       "non-targeting_02847               -1                 1                  -1   \n",
       "non-targeting_02969               -1                 1                  -1   \n",
       "non-targeting_03155               -1                 1                  -1   \n",
       "non-targeting_03351               -1                 1                  -1   \n",
       "non-targeting_03447               -1                 1                   1   \n",
       "\n",
       "                     OneClassSVM  \n",
       "non-targeting_00084           -1  \n",
       "non-targeting_00202           -1  \n",
       "non-targeting_00303           -1  \n",
       "non-targeting_00362           -1  \n",
       "non-targeting_00373           -1  \n",
       "non-targeting_00715           -1  \n",
       "non-targeting_00759           -1  \n",
       "non-targeting_00790           -1  \n",
       "non-targeting_00850           -1  \n",
       "non-targeting_00867           -1  \n",
       "non-targeting_00949           -1  \n",
       "non-targeting_01357           -1  \n",
       "non-targeting_01500           -1  \n",
       "non-targeting_01651           -1  \n",
       "non-targeting_01660           -1  \n",
       "non-targeting_01777           -1  \n",
       "non-targeting_01815           -1  \n",
       "non-targeting_01967           -1  \n",
       "non-targeting_02135           -1  \n",
       "non-targeting_02295           -1  \n",
       "non-targeting_02541           -1  \n",
       "non-targeting_02771           -1  \n",
       "non-targeting_02790           -1  \n",
       "non-targeting_02839           -1  \n",
       "non-targeting_02847           -1  \n",
       "non-targeting_02969           -1  \n",
       "non-targeting_03155           -1  \n",
       "non-targeting_03351           -1  \n",
       "non-targeting_03447           -1  "
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "apu",
   "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": "3.12.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}