plotting.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 139,
   "id": "c0d481ac-4967-4d2a-940d-158fb125191e",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.colors as mcolors\n",
    "import seaborn as sns\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "\n",
    "# Change the font globally to Times New Roman\n",
    "matplotlib.rcParams['font.family'] = 'Times New Roman'\n",
    "\n",
    "# Data initialization\n",
    "lphas = [\"0\", \"1e-4\", \"1e-3\", \"1e-2\", \"1e-1\", \"1\"]\n",
    "alphas_n = [0, 1e-4, 1e-3, 1e-2, 1e-1, 1]\n",
    "models = [\"UVA\", \"LP\", \"LPSC\", \"MNODE\", \"BNODE\",\"Tran\", \"TCN\",\"LSTM\",\"S4D\"]\n",
    "labels = models\n",
    "\n",
    "\n",
    "# Custom colors for the bars\n",
    "colors = ['#b3cde3',  # Very light blue for UVA\n",
    "          '#8c96c6', '#8c96c6',  # Lighter blue for LP and LPSC\n",
    "          '#8856a7',  # Medium blue for MNODE\n",
    "          '#810f7c', '#810f7c', '#810f7c', '#810f7c', '#810f7c']  # Darker blue for the rest\n",
    "\n",
    "data_pred={}\n",
    "data_rank={}\n",
    "for mod in models:\n",
    "    mod2=mod\n",
    "    #if mod==\"MNODE\":\n",
    "    #    mod2=mod+\"2\"\n",
    "\n",
    "    if mod==\"LP\":\n",
    "        mod2=\"LP_reduced\"\n",
    "    if mod==\"LPSC\":\n",
    "        mod2=\"LPSC_reduced\"\n",
    "    if mod==\"UVA\":\n",
    "        mod2=\"UVA_new_allrep\"\n",
    "    pred=[]\n",
    "    causal=[]\n",
    "    for a in alphas_n:\n",
    "        if True:\n",
    "            if mod==\"UVA\":\n",
    "                rmse=np.load(f\"{mod2}_a{a}_pred.npy\")\n",
    "                er=np.load(f\"{mod2}_a{a}_causal.npy\")\n",
    "            if mod=='MNODE':\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"MNODE_comp1_a{a}_pred.npy\"),np.load(f\"MNODE_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"MNODE_comp1_a{a}_causal.npy\"),np.load(f\"MNODE_comp2_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"MNODEcomp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"MNODEcomp_a{a}_causal.npy\")])\n",
    "            if mod==\"S4D\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"S4D_comp1_a{a}_pred.npy\"),np.load(f\"S4D_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"S4D_comp1_a{a}_causal.npy\"),np.load(f\"S4D_comp2_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"S4D_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"S4D_comp_a{a}_causal.npy\")])\n",
    "            if mod==\"TCN\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"TCN_comp1_a{a}_pred.npy\"),np.load(f\"TCN_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"TCN_comp1_a{a}_causal.npy\"),np.load(f\"TCN_comp2_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"TCN_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"TCN_comp_a{a}_causal.npy\")])\n",
    "            if mod==\"LP\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LP_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LP_comp_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LP_comp1_a{a}_pred.npy\"),np.load(f\"LP_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LP_comp1_a{a}_causal.npy\"),np.load(f\"LP_comp2_a{a}_causal.npy\")])\n",
    "            if mod==\"LSTM\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LSTM_comp1_a{a}_pred.npy\"),np.load(f\"LSTM_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LSTM_comp1_a{a}_causal.npy\"),np.load(f\"LSTM_comp2_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LSTM_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LSTM_comp_a{a}_causal.npy\")])\n",
    "            if mod==\"BNODE\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"BNODE_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"BNODE_compD_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"BNODE_comp1_a{a}_pred.npy\"),np.load(f\"BNODE_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"BNODE_comp1D_a{a}_causal.npy\"),np.load(f\"BNODE_comp2D_a{a}_causal.npy\")])\n",
    "            if mod==\"LPSC\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LPSC_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LPSC_comp_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LPSC_comp1_a{a}_pred.npy\"),np.load(f\"LPSC_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LPSC_comp1_a{a}_causal.npy\"),np.load(f\"LPSC_comp2_a{a}_causal.npy\")])\n",
    "            if mod==\"Tran\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"Tran_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"Tran_comp_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"Tran_comp1_a{a}_pred.npy\"),np.load(f\"Tran_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"Tran_comp1_a{a}_causal.npy\"),np.load(f\"Tran_comp2_a{a}_causal.npy\")])\n",
    "        pred.append(f\"{np.mean(rmse)}/{np.std(rmse)}\")    \n",
    "        causal.append(f\"{np.quantile(er,0.1)}/{np.quantile(er,0.5)}/{np.quantile(er,0.9)}\")        \n",
    "    data_pred[mod]=pred\n",
    "    data_rank[mod]=causal\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "def get_values_errors(data_dict):\n",
    "    values = [[float(data_dict[model][alpha].split('/')[0]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors = [[float(data_dict[model][alpha].split('/')[1]) for alpha in range(len(alphas))] for model in models]\n",
    "    return np.array(values), np.array(errors)\n",
    "\n",
    "\n",
    "def get_rank_errors(data_dict): \n",
    "    values = [[float(data_dict[model][alpha].split('/')[1]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors1 = [[float(data_dict[model][alpha].split('/')[0]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors2 = [[float(data_dict[model][alpha].split('/')[2]) for alpha in range(len(alphas))] for model in models]\n",
    "    return np.array(values), np.array(errors1), np.array(errors2)\n",
    "\n",
    "\n",
    "pred_values, pred_errors = get_values_errors(data_pred)\n",
    "rank_values, rank_errors1, rank_errors2 = get_rank_errors(data_rank)\n",
    "\n",
    "\n",
    "# Custom colors for prediction bars in light blues\n",
    "colors_blue = ['#deebf7',   # Very light blue for UVA\n",
    "               '#c6dbef', '#c6dbef',  # Second lightest blue for LP and LPSC\n",
    "               '#9ecae1',  # Third lightest blue for MNODE\n",
    "               '#6baed6', '#6baed6', '#6baed6', '#6baed6', '#6baed6']  # Fourth lightest blue for the rest\n",
    "\n",
    "# Custom colors for ranking bars in light reds\n",
    "colors_red = ['#fee5d9',  # Very light red for UVA\n",
    "              '#fcbba1', '#fcbba1', # Second lightest red for LP and LPSC\n",
    "              '#fc9272',  # Third lightest red for MNODE\n",
    "              '#fb6a4a', '#fb6a4a', '#fb6a4a', '#fb6a4a', '#fb6a4a']  # Fourth lightest red for the rest\n",
    "\n",
    "# Now, let's create the plots\n",
    "fig, axes = plt.subplots(1, 2, figsize=(20, 6),facecolor='white')\n",
    "\n",
    "# Plot for prediction values\n",
    "axes[0].bar(models, pred_values[:,0], yerr=pred_errors[:,0], color=colors_blue, capsize=5)\n",
    "axes[0].set_title('Prediction RMSE (Pure Predictive Loss)', fontsize=28, fontweight='bold')\n",
    "axes[0].set_xticks(range(len(labels)))\n",
    "axes[0].set_xticklabels(labels, rotation=45, fontsize=28, ha=\"center\")\n",
    "axes[0].set_ylabel('RMSE', fontsize=28)\n",
    "\n",
    "# Plot for ranking values\n",
    "axes[1].bar(models, rank_values[:,0], yerr=[rank_values[:,0]-rank_errors1[:,0],\\\n",
    "                                             rank_errors2[:,0]-rank_values[:,0]], color=colors_red, capsize=5)\n",
    "axes[1].set_title('Causal Error Rate (Pure Predictive Loss)', fontsize=28, fontweight='bold')\n",
    "axes[1].set_xticks(range(len(labels)))\n",
    "axes[1].set_xticklabels(labels, rotation=45, fontsize=28, ha=\"center\")\n",
    "axes[1].set_ylabel('Error Rate', fontsize=28)\n",
    "\n",
    "# Add a dashed light gray horizontal line at y=0.33 in the ranking subplot\n",
    "axes[1].axhline(y=0.66, color='gray', linestyle='--', linewidth=1, zorder=0)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"final_plots/main_complete2_alpha=0.pdf\", format='pdf', dpi=800)  # You can adjust the filename, format, and dpi as needed\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "id": "aa124c72-240b-4f4e-9d16-1c2da561e764",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.colors as mcolors\n",
    "import seaborn as sns\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "\n",
    "alphas = [\"0\", \"1e-4\", \"1e-3\", \"1e-2\", \"1e-1\", \"1\"]\n",
    "alphas_n = [0, 1e-4, 1e-3, 1e-2, 1e-1, 1]\n",
    "models = [\"UVA\", \"LP\", \"LPSC\", \"MNODE\", \"BNODE\",\"Tran\", \"TCN\",\"LSTM\",\"S4D\"]\n",
    "labels = models\n",
    "\n",
    "\n",
    "# Custom colors for the bars\n",
    "colors = ['#b3cde3',  # Very light blue for UVA\n",
    "          '#8c96c6', '#8c96c6',  # Lighter blue for LP and LPSC\n",
    "          '#8856a7',  # Medium blue for MNODE\n",
    "          '#810f7c', '#810f7c', '#810f7c', '#810f7c', '#810f7c']  # Darker blue for the rest\n",
    "\n",
    "data_pred={}\n",
    "data_rank={}\n",
    "plot_comp=False\n",
    "plot_compcomp=True\n",
    "for mod in models:\n",
    "    mod2=mod\n",
    "    #if mod==\"MNODE\":\n",
    "    #    mod2=mod+\"2\"\n",
    "    if mod==\"UVA\":\n",
    "        mod2=\"UVA_full\"\n",
    "    if mod==\"LP\":\n",
    "        mod2=\"LP_reduced\"\n",
    "    if mod==\"LPSC\":\n",
    "        mod2=\"LPSC_reduced\"\n",
    "    if mod==\"UVA\":\n",
    "        mod2=\"UVA_new_allrep\"\n",
    "    pred=[]\n",
    "    causal=[]\n",
    "    for a in alphas_n:\n",
    "        rmse=np.load(f\"{mod2}_a{a}_pred.npy\")[:6]\n",
    "        er=np.load(f\"{mod2}_a{a}_causal.npy\")[:6]\n",
    "        if plot_comp and (a==1):\n",
    "            if mod==\"LPSC\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LPSC_comp1_a{a}_pred.npy\"),np.load(f\"LPSC_comp2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LPSC_comp1_a{a}_causal.npy\"),np.load(f\"LPSC_comp2_a{a}_causal.npy\")])\n",
    "            if mod==\"LP\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LP_comp1_a{a}_pred.npy\"),np.load(f\"LP_comp2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LP_comp1_a{a}_causal.npy\"),np.load(f\"LP_comp2_a{a}_causal.npy\")])\n",
    "            if mod==\"Tran\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"Tran_comp1_a{a}_pred.npy\"),np.load(f\"Tran_comp2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"Tran_comp1_a{a}_causal.npy\"),np.load(f\"Tran_comp2_a{a}_causal.npy\")])\n",
    "            if mod==\"BNODE\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"BNODE_comp1_a{a}_pred.npy\"),np.load(f\"BNODE_comp2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"BNODE_comp1D_a{a}_causal.npy\"),np.load(f\"BNODE_comp2D_a{a}_causal.npy\")])\n",
    "            if mod==\"MNODE\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"MNODEcomp_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"MNODEcomp_a{a}_causal.npy\")])\n",
    "            if mod==\"UVA\":\n",
    "                rmse=np.load(f\"{mod2}_a{a}_pred.npy\")\n",
    "                er=np.load(f\"{mod2}_a{a}_causal.npy\")\n",
    "            if mod==\"LSTM\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LSTM_comp_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LSTM_comp_a{a}_causal.npy\")])\n",
    "            if mod==\"TCN\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"TCN_comp_a{a}_pred.npy\")])\n",
    "                #print(np.mean(rmse[:6]))\n",
    "                #print(np.mean(rmse))\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"TCN_comp_a{a}_causal.npy\")])\n",
    "                #print(er)\n",
    "            if mod==\"S4D\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"S4D_comp_a{a}_pred.npy\")])\n",
    "                ##print(rmse)\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"S4D_comp_a{a}_causal.npy\")])\n",
    "                ##print(er)\n",
    "        if plot_compcomp:\n",
    "            if mod==\"UVA\":\n",
    "                rmse=np.load(f\"{mod2}_a{a}_pred.npy\")\n",
    "                er=np.load(f\"{mod2}_a{a}_causal.npy\")\n",
    "            if mod=='MNODE':\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"MNODE_comp1_a{a}_pred.npy\"),np.load(f\"MNODE_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"MNODE_comp1_a{a}_causal.npy\"),np.load(f\"MNODE_comp2_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"MNODEcomp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"MNODEcomp_a{a}_causal.npy\")])\n",
    "            if mod==\"S4D\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"S4D_comp1_a{a}_pred.npy\"),np.load(f\"S4D_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"S4D_comp1_a{a}_causal.npy\"),np.load(f\"S4D_comp2_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"S4D_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"S4D_comp_a{a}_causal.npy\")])\n",
    "            if mod==\"TCN\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"TCN_comp1_a{a}_pred.npy\"),np.load(f\"TCN_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"TCN_comp1_a{a}_causal.npy\"),np.load(f\"TCN_comp2_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"TCN_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"TCN_comp_a{a}_causal.npy\")])\n",
    "            if mod==\"LP\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LP_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LP_comp_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LP_comp1_a{a}_pred.npy\"),np.load(f\"LP_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LP_comp1_a{a}_causal.npy\"),np.load(f\"LP_comp2_a{a}_causal.npy\")])\n",
    "            if mod==\"LSTM\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LSTM_comp1_a{a}_pred.npy\"),np.load(f\"LSTM_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LSTM_comp1_a{a}_causal.npy\"),np.load(f\"LSTM_comp2_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LSTM_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LSTM_comp_a{a}_causal.npy\")])\n",
    "            if mod==\"BNODE\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"BNODE_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"BNODE_compD_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"BNODE_comp1_a{a}_pred.npy\"),np.load(f\"BNODE_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"BNODE_comp1D_a{a}_causal.npy\"),np.load(f\"BNODE_comp2D_a{a}_causal.npy\")])\n",
    "            if mod==\"LPSC\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LPSC_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LPSC_comp_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LPSC_comp1_a{a}_pred.npy\"),np.load(f\"LPSC_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LPSC_comp1_a{a}_causal.npy\"),np.load(f\"LPSC_comp2_a{a}_causal.npy\")])\n",
    "            if mod==\"Tran\":\n",
    "                if a<1:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"Tran_comp_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"Tran_comp_a{a}_causal.npy\")])\n",
    "                else:\n",
    "                    rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"Tran_comp1_a{a}_pred.npy\"),np.load(f\"Tran_comp2_a{a}_pred.npy\")])\n",
    "                    er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"Tran_comp1_a{a}_causal.npy\"),np.load(f\"Tran_comp2_a{a}_causal.npy\")])\n",
    "                \n",
    " \n",
    "        pred.append(f\"{np.mean(rmse)}/{np.std(rmse)}\")    \n",
    "        causal.append(f\"{np.quantile(er,0.1)}/{np.quantile(er,0.5)}/{np.quantile(er,0.9)}\")  \n",
    "    data_pred[mod]=pred\n",
    "    data_rank[mod]=causal\n",
    "\n",
    "# Custom colors for the bars\n",
    "colors = ['#b3cde3', # Very light blue for UVA\n",
    "          '#8c96c6','#8c96c6', '#8c96c6',  # Lighter blue for LP and LPSC\n",
    "          '#8856a7',  # Medium blue for MNODE\n",
    "          '#810f7c', '#810f7c', '#810f7c', '#810f7c', '#810f7c']  # Darker blue for the rest\n",
    "\n",
    "colors_blue = ['#deebf7',   # Very light blue for UVA\n",
    "               '#c6dbef', '#c6dbef',  # Second lightest blue for LP and LPSC\n",
    "               '#9ecae1',   # Third lightest blue for MNODE\n",
    "               '#6baed6', '#6baed6', '#6baed6', '#6baed6', '#6baed6']  # Fourth lightest blue for the rest\n",
    "\n",
    "# Custom colors for ranking bars in light reds\n",
    "colors_red = ['#fee5d9',   # Very light red for UVA\n",
    "              '#fcbba1', '#fcbba1',  # Second lightest red for LP and LPSC\n",
    "              '#fc9272',  # Third lightest red for MNODE\n",
    "              '#fb6a4a', '#fb6a4a', '#fb6a4a', '#fb6a4a', '#fb6a4a'] \n",
    "\n",
    "def get_values_errors(data_dict):\n",
    "    values = [[float(data_dict[model][alpha].split('/')[0]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors = [[float(data_dict[model][alpha].split('/')[1]) for alpha in range(len(alphas))] for model in models]\n",
    "    return np.array(values), np.array(errors)\n",
    "\n",
    "\n",
    "def get_rank_errors(data_dict): \n",
    "    values = [[float(data_dict[model][alpha].split('/')[1]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors1 = [[float(data_dict[model][alpha].split('/')[0]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors2 = [[float(data_dict[model][alpha].split('/')[2]) for alpha in range(len(alphas))] for model in models]\n",
    "    return np.array(values), np.array(errors1), np.array(errors2)\n",
    "\n",
    "\n",
    "pred_values, pred_errors = get_values_errors(data_pred)\n",
    "rank_values, rank_errors1, rank_errors2 = get_rank_errors(data_rank)\n",
    "\n",
    "# Creating two subplots, one for data_pred and one for data_rank\n",
    "fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(12, 10),facecolor='white')\n",
    "\n",
    "bar_width = 0.1\n",
    "index = np.arange(len(models))\n",
    "\n",
    "def get_gradation_colors(base_color, n):\n",
    "    \"\"\"Generates a gradation of colors from a base color to a lighter shade.\"\"\"\n",
    "    color = mcolors.to_rgb(base_color)\n",
    "    # Adjust the factor to control the rate of color change\n",
    "    factor = 0.9  # Adjust this value as needed to control the darkness\n",
    "    return [mcolors.to_hex((color[0] + (1 - color[0]) * i / (n - 1) * factor, \n",
    "                            color[1] + (1 - color[1]) * i / (n - 1) * factor, \n",
    "                            color[2] + (1 - color[2]) * i / (n - 1) * factor)) for i in range(n)]\n",
    "\n",
    "# Generate gradation colors for each model for predictions\n",
    "gradation_colors_blue = [get_gradation_colors(color, len(alphas)) for color in colors_blue]\n",
    "\n",
    "# Generate gradation colors for each model for rankings\n",
    "gradation_colors_red = [get_gradation_colors(color, len(alphas)) for color in colors_red]\n",
    "\n",
    "# Plotting bars with gradation colors for data_pred\n",
    "for i, model in enumerate(models):\n",
    "    for j, alpha in enumerate(alphas):\n",
    "        ax1.bar(index[i] + j*bar_width, pred_values[i][j], yerr=pred_errors[i][j], \n",
    "                color=gradation_colors_blue[i][j], edgecolor='black', ecolor='#3b3b3b', \n",
    "                width=bar_width, label=f'α={alpha}' if i == 0 else \"\")\n",
    "\n",
    "# Plotting bars with gradation colors for data_rank\n",
    "for i, model in enumerate(models):\n",
    "    for j, alpha in enumerate(alphas):\n",
    "        ax2.bar(index[i] + j*bar_width, rank_values[i][j], yerr=[[rank_values[i][j]-rank_errors1[i][j]],\\\n",
    "                                                                 [rank_errors2[i][j]-rank_values[i][j]]], \n",
    "                color=gradation_colors_red[i][j], edgecolor='black', ecolor='#3b3b3b', \n",
    "                width=bar_width, label=f'α={alpha}' if i == 0 else \"\")\n",
    "\n",
    "\n",
    "# Customizing the plot for aesthetics\n",
    "ax1.set_title('Prediction RMSE (Hybrid Loss)', fontsize=26)\n",
    "ax1.set_ylim([0,100])\n",
    "ax2.set_title('Causal Error Rate 10/50/90-th Percentile (Hybrid Loss)', fontsize=26)\n",
    "ax1.set_xticks(index + bar_width * 2.5)\n",
    "ax2.set_xticks(index + bar_width * 2.5)\n",
    "ax1.set_xticklabels(models, rotation=45, fontsize=26, color='black')\n",
    "ax2.set_xticklabels(models, rotation=45, fontsize=26, color='black')\n",
    "ax1.tick_params(axis='y', labelsize=26)  # Adjust the labelsize as needed for the first subplot\n",
    "ax2.tick_params(axis='y', labelsize=26)  # Adjust the labelsize as needed for the second subplot\n",
    "\n",
    "# Add a dashed light gray horizontal line at y=0.33 in the ranking subplot\n",
    "ax2.axhline(y=0.66, color='gray', linestyle='--', linewidth=1, zorder=0)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"final_plots/main_comp2.pdf\", format='pdf', dpi=800)  # You can adjust the filename, format, and dpi as needed\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "bc7ca735-2983-4844-9cc3-1de974ce59c8",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.colors as mcolors\n",
    "import seaborn as sns\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "\n",
    "alphas = [\"0\",\"1e-4\",\"1e-3\",\"1e-2\",\"1e-1\",\"1\"]\n",
    "alphas_n = [0,1e-4,1e-3,1e-2,1e-1,1]\n",
    "#alphas = [\"0\",\"1e-2\",\"1\"]\n",
    "#alphas_n = [0,1e-2,1]\n",
    "models = [\"LP\",\"LP_full\",\"MNODE\",\"MNODE_full\"]\n",
    "labels = models\n",
    "\n",
    "\n",
    "data_pred={}\n",
    "data_rank={}\n",
    "for mod in models:\n",
    "    mod2=mod\n",
    "\n",
    "    pred=[]\n",
    "    causal=[]\n",
    "    for a in alphas_n:\n",
    "        if mod==\"LP_full\":\n",
    "            rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"{mod2}_rep1_a{a}_pred.npy\"),np.load(f\"{mod2}_rep2_a{a}_pred.npy\")])\n",
    "            er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"{mod2}_rep1_a{a}_causal.npy\"),np.load(f\"{mod2}_rep2_a{a}_causal.npy\")])\n",
    "        if mod==\"MNODE_full\":\n",
    "            rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"{mod2}_rep1_full_a{a}_pred.npy\"),np.load(f\"{mod2}_rep2_full_a{a}_pred.npy\")])\n",
    "            er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"{mod2}_rep1_full_a{a}_causal.npy\"),np.load(f\"{mod2}_rep2_full_a{a}_causal.npy\")])\n",
    "        if mod==\"LP\":\n",
    "            mod2=\"LP_reduced\"\n",
    "            if a<1:\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LP_comp_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LP_comp_a{a}_causal.npy\")])\n",
    "            else:\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LP_comp1_a{a}_pred.npy\"),np.load(f\"LP_comp2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LP_comp1_a{a}_causal.npy\"),np.load(f\"LP_comp2_a{a}_causal.npy\")])\n",
    "        if mod=='MNODE':\n",
    "            if a<1:\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"MNODE_comp1_a{a}_pred.npy\"),np.load(f\"MNODE_comp2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"MNODE_comp1_a{a}_causal.npy\"),np.load(f\"MNODE_comp2_a{a}_causal.npy\")])\n",
    "            else:\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"MNODEcomp_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"MNODEcomp_a{a}_causal.npy\")])\n",
    "                \n",
    "        pred.append(f\"{np.mean(rmse)}/{np.std(rmse)}\")    \n",
    "        causal.append(f\"{np.quantile(er,0.1)}/{np.quantile(er,0.5)}/{np.quantile(er,0.9)}\")  \n",
    "    data_pred[mod]=pred\n",
    "    data_rank[mod]=causal\n",
    "    #print(pred)\n",
    "    #print(causal)\n",
    "\n",
    "# Custom colors for the bars\n",
    "colors = ['#b3cde3', # Very light blue for UVA\n",
    "          '#8c96c6','#8c96c6', '#8c96c6',  # Lighter blue for LP and LPSC\n",
    "          '#8856a7',  # Medium blue for MNODE\n",
    "          '#810f7c', '#810f7c', '#810f7c', '#810f7c', '#810f7c']  # Darker blue for the rest\n",
    "\n",
    "colors_blue = ['#deebf7',   # Very light blue for UVA\n",
    "               '#c6dbef', '#c6dbef',  # Second lightest blue for LP and LPSC\n",
    "               '#9ecae1',   # Third lightest blue for MNODE\n",
    "               '#6baed6', '#6baed6', '#6baed6', '#6baed6', '#6baed6']  # Fourth lightest blue for the rest\n",
    "\n",
    "# Custom colors for ranking bars in light reds\n",
    "colors_red = ['#fee5d9',   # Very light red for UVA\n",
    "              '#fcbba1', '#fcbba1',  # Second lightest red for LP and LPSC\n",
    "              '#fc9272',  # Third lightest red for MNODE\n",
    "              '#fb6a4a', '#fb6a4a', '#fb6a4a', '#fb6a4a', '#fb6a4a'] \n",
    "\n",
    "\n",
    "def get_values_errors(data_dict):\n",
    "    values = [[float(data_dict[model][alpha].split('/')[0]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors = [[float(data_dict[model][alpha].split('/')[1]) for alpha in range(len(alphas))] for model in models]\n",
    "    return np.array(values), np.array(errors)\n",
    "\n",
    "\n",
    "def get_rank_errors(data_dict): \n",
    "    values = [[float(data_dict[model][alpha].split('/')[1]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors1 = [[float(data_dict[model][alpha].split('/')[0]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors2 = [[float(data_dict[model][alpha].split('/')[2]) for alpha in range(len(alphas))] for model in models]\n",
    "    return np.array(values), np.array(errors1), np.array(errors2)\n",
    "\n",
    "\n",
    "pred_values, pred_errors = get_values_errors(data_pred)\n",
    "rank_values, rank_errors1, rank_errors2 = get_rank_errors(data_rank)\n",
    "\n",
    "# Creating two subplots, one for data_pred and one for data_rank\n",
    "fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(12, 10),facecolor='white')\n",
    "\n",
    "bar_width = 0.1\n",
    "index = np.arange(len(models))\n",
    "\n",
    "def get_gradation_colors(base_color, n):\n",
    "    \"\"\"Generates a gradation of colors from a base color to a lighter shade.\"\"\"\n",
    "    color = mcolors.to_rgb(base_color)\n",
    "    # Adjust the factor to control the rate of color change\n",
    "    factor = 0.9  # Adjust this value as needed to control the darkness\n",
    "    return [mcolors.to_hex((color[0] + (1 - color[0]) * i / (n - 1) * factor, \n",
    "                            color[1] + (1 - color[1]) * i / (n - 1) * factor, \n",
    "                            color[2] + (1 - color[2]) * i / (n - 1) * factor)) for i in range(n)]\n",
    "\n",
    "# Generate gradation colors for each model for predictions\n",
    "gradation_colors_blue = [get_gradation_colors(color, len(alphas)) for color in colors_blue]\n",
    "\n",
    "# Generate gradation colors for each model for rankings\n",
    "gradation_colors_red = [get_gradation_colors(color, len(alphas)) for color in colors_red]\n",
    "\n",
    "# Plotting bars with gradation colors for data_pred\n",
    "for i, model in enumerate(models):\n",
    "    for j, alpha in enumerate(alphas):\n",
    "        ax1.bar(index[i] + j*bar_width, pred_values[i][j], yerr=pred_errors[i][j], \n",
    "                color=gradation_colors_blue[i][j], edgecolor='black', ecolor='#3b3b3b', \n",
    "                width=bar_width, label=f'α={alpha}' if i == 0 else \"\")\n",
    "\n",
    "# Plotting bars with gradation colors for data_rank\n",
    "for i, model in enumerate(models):\n",
    "    for j, alpha in enumerate(alphas):\n",
    "        ax2.bar(index[i] + j*bar_width, rank_values[i][j], yerr=[[rank_values[i][j]-rank_errors1[i][j]],\\\n",
    "                                                                 [rank_errors2[i][j]-rank_values[i][j]]], \n",
    "                color=gradation_colors_red[i][j], edgecolor='black', ecolor='#3b3b3b', \n",
    "                width=bar_width, label=f'α={alpha}' if i == 0 else \"\")\n",
    "\n",
    "\n",
    "# Customizing the plot for aesthetics\n",
    "ax1.set_title('Prediction RMSE (Hybrid Loss)', fontsize=26)\n",
    "ax1.set_ylim([0,100])\n",
    "ax2.set_ylim([0,1])\n",
    "ax2.set_title('Causal Error Rate 10/50/90-th Percentile (Hybrid Loss)', fontsize=26)\n",
    "ax1.set_xticks(index + bar_width * 2.5)\n",
    "ax2.set_xticks(index + bar_width * 2.5)\n",
    "ax1.set_xticklabels(models, rotation=45, fontsize=26, color='black')\n",
    "ax2.set_xticklabels(models, rotation=45, fontsize=26, color='black')\n",
    "ax1.tick_params(axis='y', labelsize=26)  # Adjust the labelsize as needed for the first subplot\n",
    "ax2.tick_params(axis='y', labelsize=26)  # Adjust the labelsize as needed for the second subplot\n",
    "\n",
    "# Add a dashed light gray horizontal line at y=0.33 in the ranking subplot\n",
    "ax2.axhline(y=0.66, color='gray', linestyle='--', linewidth=1, zorder=0)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"final_plots/fullmodel_complete2.pdf\", format='pdf', dpi=800)  # You can adjust the filename, format, and dpi as needed\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "da9bfd37-74b8-4087-9be8-5800dc5b0988",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1IAAALICAYAAAB1rMGWAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/OQEPoAAAACXBIWXMAAAsTAAALEwEAmpwYAACd90lEQVR4nOzdd3gUVfv/8c8mgUAaCaEGAglFQGlCpFpARAVFEEVUeARBUYGAvSBFQZoCKqBfaYrYUEDBiOiDKIiIdKSJ9C4QQodA2vn9kd/Ok2U3ZUg2CfJ+Xdde12bOmTP3zO5O5p5yjsMYYwQAAAAAyDGfgg4AAAAAAK40JFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAADko7S0tIIOAYUAnSYDVz4SKQBXvZSUFH399de6/fbbdeutt7qVjxgxQiEhIfr000+9Gkd+LQcF5+OPP9aoUaMKOgxJ0vr169WnTx+FhoZqz549edr2jBkzFBISopEjR+Z4ns2bNys2NtYr8RRGR44cUd++fbVz586CDgXAZSKRAuA1gwcPVnh4uBwOh9srKChIFSpU0O23364xY8bozJkzBRLjwYMH9cQTT6hnz55auHChx6sFn3/+uc6cOaOZM2d6NZb8Wo4dBw8e1ODBg9WsWTO3z7Bo0aIKDAxUZGSkWrVqpQkTJuj8+fMe27nzzjtVtGhRl/m/+eYbW7HcdtttLvNHRETol19+camzadMmde3aVRUrVlRUVJSioqLUvn17TZkyRb///ru6du3qtm4tW7a02ixSpIiCgoJcXgEBAfLx8bHqXM5BfmJioh5++GElJiZqwIABWr9+vSpXruzSrnP50dHR2rdvnzVvly5dFBIS4rb9w8PDNXjwYNuxSNLrr7+unj176v3339epU6cuq42sfPXVVzpz5ow+++yzHNUfNWqUunbtqokTJ+Y4nlq1asnf399tuwQHB6tz5865CT9flCtXToMHD1a3bt307bffFnQ4AC6HAQAvGzp0qJFkJJkHH3zQvPPOO2bUqFGmTZs21vQKFSqYDRs2FFiMb7zxhpFkbrnlFreyL774wtxwww3m559/zpNlZdZOXi8nL6WkpJjy5csbSaZjx47m+PHjxhhjDhw4YCZMmGBKlChhJJnrrrvO7N+/32Mbx48fN02bNrU+85iYmBwvf+XKldZ8ksySJUvc6sydO9f4+/ubJ554wpw8edIYY8yFCxfM9OnTTalSpYwk06pVK7f50tLSTMWKFY0k07NnT4/Lv3jxopk1a5YpVqyY2b17d47jds575513mtdff92tbMuWLSYwMNBIMpGRkebgwYMe20hKSjK9evWy1n/EiBEmLS3NVhyXOnjwoNWe3XXKzs8//2waNWpkPv/88xzPc+DAAdvxHD9+3NStW9ea7+OPPzapqamXGXXBOHLkiKlYsaKZPXt2QYcCwCYSKQBet3nzZutAZ+nSpS5ls2bNMg6Hw0gy0dHRJjExsUBinDx5cqaJVF76+eefTbdu3by6DG9p3rx5psnGsmXLrM/RU7Li9O6777okRD/++GOOln3vvfda85QrV86t/MiRIyY4ONjcdNNNHufftWuXKVmyZKaxZbVuGfXu3dt20tGvXz9TvXp1k5yc7LH8hhtuMJLMQw89lGU7P/74o7UN/vrrL1sxeJKcnOy1ROpyJCUlXVY8zz33nJFkSpcu7b3gvOzbb781QUFBZtOmTQUdCgAbuLUPgNcFBARkWnb//ferY8eOkqTdu3fru+++y6+wXBQpUsTry4iPj9cTTzzh9eV4i5+fX6ZlzZo1U/PmzSVJixYt0oEDBzzWCwkJUb169RQSEiJJGj58eLbL3bp1q+bNm6ebb75ZkhQYGOhWZ968eTpz5oxuuOEGj21ER0dr6NChmS4jq3XLqHv37goODs5RXUlavHixJkyYoJdffjnTZTh/H1n9Ti4tL1asWI5jyExO1zm/XO5vMCgoSJKs79SVqF27dqpRo4YeeughOqEAriAkUgAK3C233GK9/7c+eH306FHddddd2r59e0GH4jWRkZHW+8wSKUkKDQ3VU089JUn69ddftWzZsizbHT16tBo1aqSWLVtmWuf48eOSpLlz52b6vN1//vMf+fr6Zrms7Nxwww0KDw/Pcf2XX35ZPj4+at++fa6Wi3+/Ll26aOPGjYXqGUkAWSORAlDgnAfBkhQRESEpvYvouXPnqmXLlnr00UeVkpKiPn36qESJEnrggQdcOoXYtGmTHnroIdWtW1dBQUG65pprNHz4cCUnJ3tc3s8//6w777xTVatWVfny5dWuXTvt2rXLY92EhAS98847qlOnjl577TWPddasWaNOnTqpRo0aKlu2rOrUqaNx48YpKSlJUnpvZO3bt9eOHTskSV9//bWqVaumatWqaf78+TlezpYtW9SjRw/Vq1dPFSpUUHR0tHr37q2DBw+61T1x4oTeeOMNlS5dWnv27NGFCxf08ssvq1y5cipZsqSee+45paamelzO5dqyZYuk9CsLNWrUyLLuM888Y11Vyeqq1P79+/XZZ59pwIABWbZXr149SdKuXbt08803a+vWrW51QkJCVLNmzSzbyYwxRrGxsbbmWbJkiVasWKFGjRrZSr7suvfee106W/D19VW3bt1c6txzzz1WeWZXvrZt26Z77rlHwcHBKleunGJjY3Xy5EmXOmfPntXEiRNVo0YNTZ8+Xf/884/uuOMOhYSEaNy4cVadqVOnqmnTpnr00Uc9Luvo0aPq27evqlevrujoaF177bUFlkCcP39eo0eP1g033KAaNWqoTJkyatOmjRYsWOCx/t69e9WpUyfVqlVLJUqUsLbrpb/bnNZzcp5QGjZsWF6uHgBvKuh7CwH8++3evTvTZ6QSExPNtddeaySZsLAwc/z4cfP777+7dErQrVs306dPH6tDA0lmx44dxhhjZs+ebWrUqGFWrVpljDHmn3/+MTfffLORZNq1a+f2QP64ceOMj4+PGT9+vElNTTVJSUlm0KBBVrsZn5H673//azp16mQ9+zNkyBC3dZs0aZIpW7as+e9//2uMSe/c4M477zSSzB133OGy/CFDhljrk1FOljNnzhwTFBRkpk6dalJTU01aWpqZPXu2KV68uClRooRZvny5VXfUqFGmQoUK1jpt3LjRNGvWzISFhZmSJUta00eNGpXtZ5fRLbfckulzRNOnT7fafe655zJt46OPPrK2cZ8+fax51qxZ47F+//79TZ06dUxaWpq1/apWrepWLzU11dx0001We0WLFjUvv/yy1elEbtYtKSnJjB071vazbc7OIZ544onLXnZGS5cu9fgMUUpKinnhhRessokTJ3qc/5577jF16tQxp0+ftqY55/n6669NaGioiYyMtDq/kORSf9q0aaZKlSpW2aRJk0zjxo1NQECAkWQqVqxo1qxZY7p06WKKFSvm8btujDF//fWXKVu2rGndurWJj4+31s3ZIcil65edrL4X2Tl06JCpV6+ead++vUlISDDGpD9P53xu7cUXX3Spf/r0aRMVFWXeeecda9o333xjAgMDXX63Oa2XUXJysvHz8zOSzJYtW2yvC4D8RyIFwOsyJlJxcXHW9C1btpjWrVsbSSYwMNAsWLDAZb5OnToZSebaa681b7/9tklMTDSDBg0yzz77rElLSzM7d+40AQEBbj24HThwwEpKZsyYYU1ftGiRcTgc5vHHH3eL0XkQ7qmziVatWnlMcH7++Wfj4+Pj1tvW559/bq3v2rVrremZJVLZLWf79u0mMDDQ9OjRw22e9957z3rQ/ujRo8aY9F7itm7dasXQtm1bM2fOHJOWlmbS0tLMI488Ym1XOy494E9JSTHr1q0zffv2Nb6+vqZIkSLmpZdeMikpKZm2kTGR2rNnj3XgeN9997nVjY+PNwEBAeazzz4zxmR/wHz8+HErRuerZMmS5s0338y2ExPnfCVKlDA1atSwXtHR0VaMdhOp6OhoI8mMGzcuR8u+3ETKmPTPwtl73WuvveZx/po1a5ovv/zSZZqzvVtvvdXq6CAxMdE8++yzVlm/fv2s+klJSSYiIsJIMg0aNDC///672b9/v+nZs6f58MMPrXo9e/b0uM0SExPNNddcY8qUKWNOnDjhUjZt2rR8T6Ruv/12Exoaak6dOuUy/dChQyYkJMRIMtOnT7emf/LJJ0aSSzJqTPrJi4y/25zWu5TzO/Pmm2/aXhcA+Y9ECoDXZUyk7rzzTtOiRQtTq1YtU758eXP99debp59+2uzcudNtvgEDBhhJplGjRh67eu7fv7+JiIjwuMyyZcsaSeauu+6yptWpU8e6QnOpsWPHZppIPfzwwx4TnKZNm5qSJUu6dbd8+vRp061bN/PAAw+YM2fOWNOzS6QyW06XLl2MJLdE05j0s9hlypQxkswrr7xiTb948aK1zVeuXOkyz6+//mokmYCAAI9xZMZ5wB8cHGyqVatmgoODrWUMHz7cHDp0KNs2MiZSxhjTrVs3I8k4HA63s/CDBg0yVapUsRKznBwwp6ammnfffdfl6qWU3rX4119/ne26eUpmTp48af7zn//YSqSOHz9uLXvmzJlZ1s0sibv0ValSpSwTja+//tpI6UMJXLx40aVsxYoVpkyZMiYpKcll+qVXeDPq0KGDkWSCgoJcehxs1qyZkWReeOGFTNfJ+du9dJuNGTMm06uWCQkJ+ZpILVy40EgynTt39lj+zDPPWN8d53fQGf/QoUNd6h49etRlWk7rXaply5ZGkunUqZOtdQFQMHhGCkC+evXVV/XLL79oy5YtOnTokNauXau3335bVapUcavr7MWrVq1acjgcbuULFy7UiRMnVLNmTbeXMUbh4eE6ceKEpPTnmDZu3KigoCBdd911bm2VLFky05g99SYWHx+vP/74Q9dcc418fFx3pcHBwZo+fbq+/PJLq0exnPC0nPPnz2vOnDmS0nueu5Sfn5/atGkjKb3nOk9tlS5d2mWe8uXLW21fjmbNmmn79u1atmyZ9ZzT2rVrrXbtePnll+VwOGSM0ciRI63pZ8+e1XvvvacXX3zRVgcRPj4+6tevn7Zv364+ffpY22H//v3q2LFjts9aeVKiRAmNGTPG43cwMxmfW8vpd+D+++/X1q1bM31lN7hthw4dVKtWLR08eFAff/yxS9mkSZPUvXv3THvG87SN+/fvLyn9s9i2bZs13dnGtddem2ksmS1n+vTpkqTGjRu7lWX1G/SGTz75RJLn35WU/lyZlP7dWbt2rSRZHZ4MHjxYd955p1atWiUp/Tc2aNAga96c1ruUs0fIw4cPX/Z6Acg/JFIArlh79uxR7dq1PR50HjlyRMeOHbN6hFuxYoUkKTw83NYBcWZ2794tY4xLpxfesGvXLl24cEGSVLRoUY91nInh3r17rWlZrWNedXvt7FRDkubMmaPx48fbbqNmzZq69957JUlffPGF9uzZIyn9wN/f31/du3e/rNhKly6tiRMnasuWLbrvvvus6SNHjlRcXJzt9sqUKaMyZcrkuP7p06et98WLF7e9vMvhcDj08ssvS0rv6dDZmcipU6f05Zdf6vHHH7fV3vXXX2+9v7TTictx7tw5bdq0SZJ7cl8QNm/eLCn735X0v99WgwYNNHnyZBUvXlw//vijGjVqpNatW1uJllNO613K2RHI0aNHL3u9AOQfEikAV6yUlBRt3749R8mM88pUZj352WX+/1gv27dv9+q4L+fOnbPeHzt2zGOdsLAwSelXTvLbU089ZY0D9sILL2jlypW223BeJUpJSdHo0aOVlJSkcePG6dlnn5W/v3+O2pg9e7Z+++03t+nVqlXT7NmzNWPGDCu5fP/9923HKKUnJzmV8SpUYmLiZS3vcjz88MOqXLmydu7cqS+//FKS9Omnn6pJkyaqVq2arbYyjtfl/I7lRsZkLK9+h7nh/G1l97uSXH9bjz/+uLZu3aoePXrIz89PP/30k2644QZNmjTJZf6c1svIeXU7s+QOQOFCIgXgilWhQgWdPHlS33//faZ1/vvf/0r6321DR44cyZMD24oVK0pKT9AWLVrksc7p06ezvR0rOxlveczsbHZKSoqk/3UBnt+mTp2qSpUqKSkpSZ07d7aS1pxq2LChbr/9dknSRx99pFGjRikxMVFPPvmkrXY8JVJO//nPf/TII49ISr+amBvJycnZjndWrlw56/3l3kJ5Ofz8/PT8889LkkaNGiVjjCZNmqRevXrZbss5LEFISIjtJMyTsLAwK5l1XnksCOvWrZMkVa1aVVL2vysfHx/Vrl3bpaxSpUqaNm2aNm/erJYtWyotLU2xsbFu65XTek7OMdDy+zZHAJeHRArAFcv5HEKfPn08jqW0bt06a2ya5s2bS5JSU1P1yy+/ZNpmTq8uVahQwRor6eWXX7bGjMpowIABLonQ5dxSWLp0ad10002SZD0rdSnnbUcPP/yw7fbzQlhYmD7//HP5+vpqz549euSRR2xfpXNelbp48aKGDBmi2NhYW8+XSelXXrISExMjSR6fx7Nj/PjxOnToUJZ1SpUqZT3vkt+3afXs2VNlypTRxo0b9corr+jIkSPW7ZN2OK8uPvjgg5k+82RHQECA6tevLyn9+caseOsq744dO6zfv/NK6sqVK7V//363us7fVcuWLa3n/8aOHesyRtk111yjhQsXKiYmRsnJydY2y2m9S506dUqSPD7HCaDwIZEC4HUZrwCdPXs2x/M5b9nL7Iz+s88+qyJFimjfvn2KiYnRBx98oF27dmnnzp2aMmWK2rRpo379+kmS6tatqxYtWkhKH/DSebY5q1idnEnSpbcjvfDCC5LSO7K4++67tXHjRqWkpOjvv/9Wjx49tHHjRjVt2tSq73xWxvnMU06XM3z4cDkcDi1cuFBLly51KTPG6LvvvlODBg300EMPWdMzLiOzdfW0rKw4PwdPn0fz5s2tgUa/++4766rIpc6ePetx/W+55RY1a9ZMUvotZc7P7dJ5pcxvldu8ebPeeOONTOP/6aefJKUn3pdybqPsbhPdvHmzpk2bphtvvDHLej4+PrrtttskKdPBnp2c2zO7K6UZt3tWV7mKFy+up59+WlL67YhZdTLh5ClxeeeddxQREaGhQ4e6TM/udyll/l12dmDxzTffaP369ZnOb+eqsTOerL7nznpPPvmk9cxc165ddd111yktLU2vv/66W/24uDj5+vpq1KhR1jRjjNstnr6+vrr55psl/e9KdU7rXWr79u2SpCZNmmS5LgAKicvp6u/33383HTt29DimyaW++uor06hRIxMdHW1q1Khhhg8f7tb9al7MA6DwGjx4sNWtcffu3bMcZ8gpKSnJ3HHHHdYYSZ66RzfGmA8//ND4+vq6dHXtfI0fP96l7p49e0xkZKSRZDp27GgNBrpp0yara/SgoCCzfPlya0yds2fPmqpVqxpJ5sYbb3TpBtoYYx599FGPy65UqZLZs2ePS90FCxYYSaZKlSomMTHRbNmyxXz++ec5Ws6ECROMw+Ew5cuXN7///rsxxpjz58+bJ554wlSpUsVs27bNpf6PP/5oxZJxHBxj0gcxdpY5BxLOztq1a61BVoODgz12l52ammpatGhhtX3//febdevWWV3XJyYmmttuu80EBga6xWuMMd99952RZJ555hm3shMnTlifkcPhsLaB06xZs6zlPvzwwy5d3CckJJjY2FjjcDjMG2+84dZ2WlqaqVy5spFkbrrpJo/rn5aWZn7++WdTsWLFLAcczmjq1KluXfBfasuWLdaAtpUrVzYHDhzwWC8lJcX06NHDWsfXXnvNrdv9jE6ePGlCQkKMw+Ew27dvz7Sec4DdG2+80WzevNkYk/69euqpp0xUVJTbUAFHjx61utu/6aabzLlz5zzG2rx5cyPJVKtWzWUIAGOMNY5ZhQoVzK+//mqMSf/+x8bGWus3bNgw89tvv2Uad0bOgY+LFCli9u/f77HO1q1bTdu2bU2jRo1cpm/bts2UL1/e6sI/KSnJpKWlmXnz5pmwsDAzadIkl/pvvfWWNbbW8ePHjTHG7N2710RFRZm7777bdr2MTp48aSQZf3//HA8kDaBg2Uqk5s6da2699VZrR5fdmBp9+/Y1xYsXt/5R79u3z1SpUsW0bt3aXLhwIc/mAVA4DRo0yISGhrolGcHBwebhhx/OdL7vv//eBAUFuczj5+dn+vfv77H+smXLTJs2bUxISIgpXry4adasmcvAvxn9888/5rHHHjOlS5c2gYGBplWrVmbIkCFm/PjxJjw83Nx3333mvffeMzt37jSfffaZy1hJkkxoaKjLAMBpaWlmypQppl69esbf39+UK1fO9O7d20rSLvX000+boKAg06BBA+sgLSfLMcaYJUuWmLvuusuULFnSREdHm3r16pnBgwebhIQEl3odO3Z0SS4dDoepWLGiMcaY6tWrGx8fH6vMx8fHtG/fPtPPYt26daZixYpun6Gvr6/HA/+DBw+aUqVKudStWLGiufvuu61EzLncyMhItwP1Ro0aubVZr149U6RIEZc2HQ6HqVChgvn555+NMemJ1IQJE8xPP/1kBg4caJo2bWoqV65sKlasaMqUKWM6dOhgFi9e7NLu0aNHzfvvv2/uuusul7ZLly5tqlatar0qVapkAgMDrfJVq1Zlur0yOn36tAkPDzfh4eFu46A5t6tz4OiM2zUqKsrs3bvXqvvwww+7/R4kmbCwMDNo0KBMl9+nTx/TqlWrLGO8cOGCmTx5smnZsqUJDQ01UVFRplGjRua1115zG6R20KBBpmjRoi4xBAQEuIyTtWTJEhMWFuZSJygoyBpU2Zj038x7771nrr32WlOkSBFTt25d8+ijj5p169aZIkWKmGbNmpmBAwean3/+2eP4cU5Lly41b7/9tilevLi1LH9/f1OpUiXrs4uKirIG1pVkxowZ49ZOfHy86d+/v4mKijLh4eGmdu3apnPnzm7jrxnzvwTJ+R2sVKmSqVOnjhk7dqzL2F05rZeR80RCly5dMv/AABQqthKpHTt2mJSUFGvAuKwSqSlTplhnlTJy7iief/75PJkHAIDC6s033zSSzLJly/J92TExMebLL7/M9+Xi8jz11FPG19fX/P333wUdCoAcchhj/4nOl156SW+++aa6detmDa6X0fnz5xUVFaX4+HgdPHhQERERVpkxRmXKlNHx48f1999/Wz0BXc48AAAUZomJibrhhhvUoEEDzZgxI9+Wu3btWrVp00YHDhzIk44i4F2nT59WdHS0nnzySQ0fPrygwwGQQ5fV2YRzJPvMLFiwQPHx8YqKinJJiKT0XquaNGmitLQ0TZkyJVfzAABQmBUvXlzz5s3TggULtHz58nxb7ujRo/XYY4+RRF0h3nzzTdWqVcvqsAXAleGyEqnsuvB1jhrvHKPhUnXr1pUkl7FXLmceAAAKu6pVq+qbb75R7969deDAgTxv3xijxx57TE2bNtXIkSP1/PPP68cff9Rzzz2X58tC3vv+++/1008/af78+SS+wBXGzxuNOge7q1Spksdy52jhGzZsUFpamnx8fC5rHgAArgQ33nijvvrqKz3//PN69dVXVadOnTxr+/jx45o2bZok6Y8//pCvr69mzZrFoK5XgGnTpunPP//UL7/8Yg2PAODK4ZVE6siRI5Kk0NBQj+UlSpSQlD6+xKlTpxQWFnZZ82Q0efJkTZ48WZK0detW1axZM7erAQBAnjLG6L777lNISEietlu6dGklJCSoWLFiqlChgoYPH86zNoVccnKykpKSFBgYaA26DaDw2bNnj44dO+axzCuJVEJCgqT0Ucw9yXhrYGJiosLCwi5rnox69eqlXr16SUofvX716tWXvwIAAAAArnoxMTGZlnnl/jh/f/8sy50jnktSeHj4Zc8DAAAAAAXBK4lU+fLlJUlnz571WH7y5ElJUnBwsJVAXc48AAAAAFAQvJJI1atXT5J06NAhj+XO56Hq16+fq3kAAAAAoCB4JZFq27atJGnTpk0ey7dt2+ZS73LnAQAAAICC4JVEqlOnTgoPD9fWrVutK0lOqampWrVqlfz8/NS5c+dczQMAAAAABeGyEqmUlBRJ6QmOJ8HBwRo6dKiMMZoxY4ZLWVxcnBISEhQbG6vo6OhczQMAAAAABcF2IpWamqqVK1dKkjZv3qyLFy96rNe7d2916dJFI0aM0Jo1ayRJ69evV+/evdWmTRuNGjUqT+YBAAAAgPzmMMaYnFbu2bOnZs2apTNnzljTAgIC1KpVK3377bdu9Y0xmjBhgiZNmqQLFy4oNDRUvXv3Vvfu3eXr6+txGZczz6UYRwoAAABAbmWVV9hKpK4UJFIAAAAAciurvMIrnU0AAAAAwL8ZiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2eT2R+vPPP9W5c2dFRESocuXKqlixojp37qyNGzdmOs+sWbPUuHFjValSRTVr1tSIESOUnJzs7VABAAAAIEe8mkh9//33atSokYoVK6bNmzdr7969+vvvvxUWFqYbbrhBCxcudJsnNjZW3bp10xtvvKFdu3Zp4cKFmjZtmu666y5dvHjRm+ECAAAAQI44jDHGGw0nJSUpMjJSYWFh2rx5s3x9fa2ylJQU1ahRQ0lJSdq7d698fNLzualTp+rxxx/XsGHDNHDgQKv+/Pnzdffdd+v555/XW2+9le2yY2JitHr16rxfKQAAAABXjazyCq9dkdqwYYOOHj2qevXquSRRkuTn56cGDRrowIEDOnbsmCTp/PnzGjBggCSpR48eLvXbtm2rUqVKady4cdqxY4e3QgYAAACAHPFaIhUYGChJWrFihVJSUtzKT58+rcjISJUpU0aStGDBAsXHxysqKkoREREudR0Oh5o0aaK0tDRNmTLFWyEDAAAAQI54LZGqWbOmrr32Wu3du1eDBg1yKTty5IiWLl2qd955x5oWFxcnSapatarH9urWrStJWrRokXcCBgAAAIAc8loi5XA4NG3aNAUEBGjUqFHq16+fUlJSdOHCBT322GP6+OOP1bFjR6v+unXrJEmVKlXy2F5YWJik9FsG09LSvBU2AAAAAGTLz5uNN2nSRAsWLFD79u01YcIEbd68WREREZowYYKioqJc6h45ckSSFBoa6rGtEiVKSJKSk5N16tQpK7Fymjx5siZPnixJio+Pz9sVAQAAAIAMvD6O1M0336wJEyboqaee0l9//aVPP/1Uzz//vM6dO+dSLyEhQZIUEBDgsR2Hw2G9T0xMdCvv1auXVq9erdWrV6t06dJ5uAYAAAAA4MrridTUqVP1+++/6/3339eqVatUp04dzZkzRy1bttTJkyetev7+/lm2k5SUZL0PDw/3VrgAAAAAkC2vJlIzZsxQ37599cYbb0iSKlSooF9//VUNGjTQqlWr9Oijj1p1y5cvL0k6e/asx7acSVdwcHC2SRcAAAAAeJPXEqnTp0/rmWeeUbNmzVSyZElremhoqBYsWKCIiAjNnTtXK1eulCTVq1dPknTo0CGP7Tmfoapfv763QgYAAACAHPFaIrV06VIdP35cZcuWdSsrU6aMXnjhBUnS8uXLJaUPuitJmzZt8tjetm3bXOoBAAAAQEHxWiLlHIR33759HsurV68uSSpSpIgkqVOnTgoPD9fWrVutq09OqampWrVqlfz8/NS5c2dvhQwAAAAAOeK1RKply5YKCwvTqlWrtH37drfy5cuXy9/fX+3atZOU/uzT0KFDZYzRjBkzXOrGxcUpISFBsbGxio6O9lbIAAAAAJAjXkukQkJCNGPGDPn4+Oi+++7Txo0bJUlpaWmaMWOGxo0bp/fff1+RkZHWPL1791aXLl00YsQIrVmzRpK0fv169e7dW23atNGoUaO8FS4AAAAA5JhXB+S9++67tWLFCo0YMUKtW7eWw+FQ0aJF1bBhQy1atEhNmzZ1m+eTTz7RhAkT9Mgjj+jChQsKDQ3VsGHD1L17d/n6+nozXAAAAADIEYcxxhR0EHktJiZGq1evLugwAAAAAFzBssorvD4gLwAAAAD825BIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBN+ZpI/fTTT+rVq5fuu+8+Pf/88/r222/d6syaNUuNGzdWlSpVVLNmTY0YMULJycn5GSYAAAAAZMkvPxayZ88ePfroozp+/Ljee+893XjjjR7rxcbGatq0aZo3b55at26t/fv3q0WLFlq8eLHi4uLk7++fH+ECAAAAQJa8fkVq2bJlatiwoUqWLKk//vgj0yRq6tSpmjhxogYMGKDWrVtLkiIjIzV+/HgtXLhQAwcO9HaoAAAAAJAjDmOM8Vbj69atU4sWLRQTE6MffvhBRYoU8Vjv/PnzioqKUnx8vA4ePKiIiAirzBijMmXK6Pjx4/r7779VrVq1bJcbExOj1atX59l6AAAAALj6ZJVXeO2K1MmTJ9W2bVulpaXpk08+yTSJkqQFCxYoPj5eUVFRLkmUJDkcDjVp0kRpaWmaMmWKt8IFAAAAgBzzWiI1YMAAHT58WH379nVLji4VFxcnSapatarH8rp160qSFi1alLdBAgAAAMBl8EoidfjwYU2ePFmS1Lx5cz3//PO6/fbbFRkZqZiYGE2YMEFpaWlW/XXr1kmSKlWq5LG9sLAwSdKGDRtc5gMAAACAguCVROqrr75Samqq/P39tWTJEvXv31///e9/9f333+v8+fPq16+fHnzwQTkfzzpy5IgkKTQ01GN7JUqUkCQlJyfr1KlT3ggZAAAAAHLMK4nU4sWLJUkdOnTQW2+9pcjISElSnTp1NGfOHDkcDs2aNUtTp06VJCUkJEiSAgICPLbncDis94mJiR7rTJ48WTExMYqJiVF8fHxerQoAAAAAuPFKIrV//35JUoUKFdzKatWqpVtvvVWSrEQqu/GhkpKSrPfh4eEe6/Tq1UurV6/W6tWrVbp06cuKGwAAAABywiuJ1OnTpyVJISEhHsvvvvtuSdKmTZskSeXLl5cknT171mP9kydPSpKCg4MZlBcAAABAgfNKIlWmTBlJ/0uoLlWxYkVJsp6RqlevniTp0KFDHus7n6GqX79+XoYJAAAAAJfFK4lUo0aNJKX3sudJ8eLFJUk1atSQJLVt21bS/65QXWrbtm0u9QAAAACgIHklkXr44YclScuWLbNuy8to165dkqQHHnhAktSpUyeFh4dr69at1tUnp9TUVK1atUp+fn7q3LmzN8IFAAAAAFu8kkg1bNhQXbt2VWJiosaPH+9SZozR9OnTVblyZcXGxkpKf/Zp6NChMsZoxowZLvXj4uKUkJCg2NhYRUdHeyNcAAAAALDFK4mUJE2aNElNmjTRsGHD9OWXX0pK733vhRde0J49exQXF6egoCCrfu/evdWlSxeNGDFCa9askSStX79evXv3Vps2bTRq1ChvhQoAAAAAtjiMs8cHLzh//rzeeOMNff7557p48aLCwsJ00003aciQIYqIiHCrb4zRhAkTNGnSJF24cEGhoaHq3bu3unfvLl9f3xwvNyYmRqtXr87LVQEAAABwlckqr/BqIlVQSKQAAAAA5FZWeYXXbu0DAAAAgH8rEikAAAAAsIlECgAAAABsIpECAAAAAJtIpAAAAADAJhIpAAAAALCJRAoAAAAAbCKRAgAAAACbSKQAAAAAwCYSKQAAAACwiUQKAAAAAGwikQIAAACQZ/r27au+ffsWdBheRyIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAQH379lXfvn0LOgwAuGKQSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANuV7InXo0CGVKlVKLVq0yLTOrFmz1LhxY1WpUkU1a9bUiBEjlJycnH9BAlcIBtAEAAAoGH75uTBjjLp166aEhIRM68TGxmratGmaN2+eWrdurf3796tFixZavHix4uLi5O/vn48RAwAAAIC7fL0iNW7cOP3xxx+Zlk+dOlUTJ07UgAED1Lp1a0lSZGSkxo8fr4ULF2rgwIH5FSoAAAAAZCrfEqk///xT48eP16BBgzyWnz9/XgMGDJAk9ejRw6Wsbdu2KlWqlMaNG6cdO3Z4PVYAAAAAyEq+JFKJiYnq0qWLJk2apDJlyniss2DBAsXHxysqKkoREREuZQ6HQ02aNFFaWpqmTJmSHyEDAAAAQKbyJZF6/vnn1bp1a915552Z1omLi5MkVa1a1WN53bp1JUmLFi3K+wABAAAAwAavdzYxf/58/f7771k+GyVJ69atkyRVqlTJY3lYWJgkacOGDUpLS5OPDz23AwAAACgYXk2kjhw5or59++r777/Ptre9I0eOSJJCQ0M9lpcoUUKSlJycrFOnTlmJldPkyZM1efJkSVJ8fHwuIwcAAACAzHn1ss6jjz6ql19+WbVq1cq2rrNL9ICAAI/lDofDep+YmOhW3qtXL61evVqrV69W6dKlLzNiAAAAAMie165ITZgwQcWKFdMTTzyRo/r+/v5KSUnJtDwpKcl6Hx4enuv4AAAAAGQtKipKe/futTVP7dq1JbleCMmpypUra8+ePbbnKwheSaQ2b96sd999VytXrszxPOXLl9eOHTt09uxZj+UnT56UJAUHBzMoLwAAAJAP9u7dq4vJqbbmeaZ/P0nSmnV/2l6efxFf2/MUFK8kUm+//bZ27tyZ5ZWjJUuWWFnq7t27Va9ePe3YsUOHDh3yWN/5DFX9+vXzPF4AAAAAsMMriVTp0qVVo0YNj2WnTp3S4cOHVbx4cauHviJFiqht27aaM2eONm3a5HG+bdu2SUofnBcAAAAACpJXOpsYOXKktm7d6vE1cuRISVKjRo2saRUqVFCnTp0UHh6urVu3WlefnFJTU7Vq1Sr5+fmpc+fO3ggZAAAAAHKs0AzGFBwcrKFDh8oYoxkzZriUxcXFKSEhQbGxsYqOji6gCAEAAAAgXaFJpCSpd+/e6tKli0aMGKE1a9ZIktavX6/evXurTZs2GjVqVAFHCAAAAABeHpD3cnzyySeaMGGCHnnkEV24cEGhoaEaNmyYunfvLl/fK6cXDwAAAAD/XvmeSHXv3l3du3fPtNzhcKhfv37q169f/gUFAAAAADYUqlv7AAAAAOBKQCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgU74PyAsAALwrKipKe/futTVP7dq1JUkOh8P28ipXrqw9e/bYng8ArmQkUgAA/Mvs3btXF5NTbc3zTP9+kqQ16/60vTz/Ir625wGAKx239gEAAACATSRSAAAAAGATiRQAAAAA2MQzUkAhwcPhAAAAVw4SKaCQ2Lt3r1JS02zN079frCRp/Z8bbC/Pz5cL0gAAAJeLIykAAAAAsIlECgAAAABsIpECAAAAAJtIpAAAAADAJhIpAAAAALCJRAoAAAAAbCKRAgAAAACbSKQAAAAAwCYSKQAAAACwiUQKAAAAAGwikQIAAAAAm0ikAAAAAMAmEikAAAAAsMmvoAMAAAAA8O/x9rvjCzqEfMEVKQAAAACwiUQKAAAAAGwikQIAAAAAm0ikAAAAAMAmEikAAAAAsIlECgAAAABsIpECAAAAAJtIpAAAAADAJhIpAAAAALDJ64nUtGnT1LBhQxUvXlwhISFq2bKlvv3220zrp6WladKkSapbt66qVKmiunXr6oMPPpAxxtuhAgAAAECO+Hmz8ccff1xTp06VJPn7++vMmTNavHixFi9erFdeeUUjRoxwqZ+SkqJOnTrp999/14IFC9SgQQNt3LhRLVq00B9//KGPPvpIDofDmyEDAAAA/zp9+/aVJE2cOLGAI/n38NoVqW+//VbffPONpk2bpjNnzujChQtasWKFrr/+eknSyJEj9dNPP7nM89prr2nu3LkaO3asGjRoIEmqU6eOhg4dqo8//ljvv/++t8IF8kTfvn2tHRUAAAD+vbyWSE2bNk0LFy5Ujx49FBQUJElq1KiRFixYoLCwMEnSJ598YtU/ePCgxowZo6CgIHXq1Mmlra5du8rHx0evvPKKTp8+7a2QCxUOyAHkFfYnAPIK+5O8w7a88nktkWrZsqV19SmjsmXLqlu3bpKko0ePWtNnzpypixcvKiYmRv7+/i7zlChRQtdee63OnDmjzz77zFshI5/k946DHRWQv/jN5Z383JZvvzteb787Pl+WBQD/Bl5LpJ5++ulMy6655hpJUlRUlDUtLi5OklS1alWP89StW1eStGjRorwJEC448AFwJeLEDPDvlp+/OX7fsMurnU1kJj4+XpJ03333WdPWrVsnSapUqZLHeZy3AzrrAZDeHT+hoEMAAAC4KhXIOFI//vijYmJidNttt0mSLly4YD37FBoa6nGeEiVKSJKOHTuWLzECAAAAQGby/YrUihUrtHLlSi1fvtyalpCQYL0PCAjwOJ+z2/PExESP5ZMnT9bkyZMl/e+KFwAAAAB4Q74mUqmpqerbt69GjBihmJgYa/qlnUt4kpSUJEkKDw/3WN6rVy/16tVLklzaBgAAAP5NoqKitHfvXlvz1K5dW5IYkzUP5WsiNXjwYNWuXVsvvPCCy/SwsDD5+/vr4sWLOnv2rMd5T548KUkqXbq0t8MELHZ3VOykAAC4+lSOitK+fE5sTpxLtlX/5ef7S5KWrrDX30BYYBFb9a8m+ZZIffHFF9qyZYtmzZrlVubr66vatWtrzZo1OnTokMf5jxw5IkmqX7++N8P0ivz+cVWqXFl79+yxPd/lyO8zIgEBATp//ny+LU+Szl9MzXHd557pJ0laueZP28sJ8Pe1PQ+uPpUqR2n/vvz7zRUvHqDExPz5zVWqVFl79+6xNc+VhDPIKGzye38SWamy9v1Lf+P79u7VweMXbM3z2ivPSpJ+/HW17eVVKFnM9jzIe/mSSP3www/67LPPNGfOHPn5eV5k27ZttWbNGm3atMlj+bZt26x6V5p9e/dqV7znZ7syM/zV9B/Xt7+ssr28WhVDbe/gcrNjPHo6yVb9gS8+LUn6+fe1tpdVJqSoTp63eQbmufQzML+ttN/jY2gAZ2FQuOzft1e/bz9ha57/G/myJGnyN0ttL69Z9TCt32tvIPR3hr0oSfp0/u+25mt0Tel83XcVDwhQYj6fmOEMMgqT/N6ftLiuXP7+xm2eCOLEBezyeiL1448/6p133tHXX3/t8Vmo7du3q3r16urZs6dGjRql5cuXKzk5WUWK/O+fwPHjx7Vt2zaFhYXprrvu8nbIV7yLFy/qwAl7Z0Vefzk9cfthqb2zIhXDOCMCIG8kXbyo7YfP2Zpn5ODnJUnf/LTC9vKqlwvkDDKQj5KSLubbiRlJql85xNY+Jbf7E1x9vNr9eVxcnMaOHavZs2e79cZ3/vx5jR07VtOnT5ckVa5cWU8//bROnjypuXPnutSdMWOG0tLS9Nprryk4ONibIQMAAABAtrx2RWr69Ol6/PHHFRgYqKioKJeylJQUnTlzRmlpadqyZYs1/Y033tDq1av1wgsvqGHDhqpSpYoWLVqkQYMG6bHHHlNsbKy3wgUAAACAHPNKIvXFF1/o0UcflSSdOnUq03oxMTGqVauW9XfRokW1YMECjRw5UnfeeadSU1NVrlw5zZgxQ/fee683QgUAAAD+9UaNebegQ/jX8Uoi9dBDD+mhhx66rHn9/f312muv6bXXXsvboAAAAAAgj+TrOFKAJL3x5jv5urxRY/PvDMzYt8fn27KAwurpQW/m27JeGTom35ZVEDiDDPx7vTZyXEGHgFwikSqkXh3OjwtA3njqlVEFHcK/Bgc+uNrl9/6EEzMozEikIEkaMoqDAwAAACCnvNr9OQAAAAD8G5FIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhUKBOpWbNmqXHjxqpSpYpq1qypESNGKDk5uaDDAgAAAABJkl9BB3Cp2NhYTZs2TfPmzVPr1q21f/9+tWjRQosXL1ZcXJz8/f0LOkQAAAAAV7lCdUVq6tSpmjhxogYMGKDWrVtLkiIjIzV+/HgtXLhQAwcOLOAIAQAAAKAQJVLnz5/XgAEDJEk9evRwKWvbtq1KlSqlcePGaceOHQURHgAAAABYCk0itWDBAsXHxysqKkoREREuZQ6HQ02aNFFaWpqmTJlSQBECAAAAQLpCk0jFxcVJkqpWreqxvG7dupKkRYsW5VtMAAAAAOBJoUmk1q1bJ0mqVKmSx/KwsDBJ0oYNG5SWlpZvcQEAAADApRzGGFPQQUhSuXLldOTIET3zzDMaN26cW/mUKVPUq1cvSdLx48etxMpp8uTJmjx5siRp69atqlmzpveDRr6Lj49X6dKlCzoMAP8C7E8A5BX2J/9ee/bs0bFjxzyWFZruzxMSEiRJAQEBHssdDof1PjEx0S2R6tWrl5Vo4d8rJiZGq1evLugwAPwLsD8BkFfYn1ydCs2tfdmND5WUlGS9Dw8P93Y4AAAAAJCpQpNIlS9fXpJ09uxZj+UnT56UJAUHBzMoLwAAAIACVWgSqXr16kmSDh065LH8yJEjkqT69evnV0gohLh9E0BeYX8CIK+wP7k6FZpEqm3btpKkTZs2eSzftm2bSz1cndhRAcgr7E8A5BX2J1enQtNr35kzZxQdHa3jx4/rn3/+UdmyZa2y1NRUlS1bVqdOndK2bdsUHR1dgJECAAAAuNoVmitSwcHBGjp0qIwxmjFjhktZXFycEhISFBsbSxIFAAAAoMAVmitSTl27dtX8+fP1008/qWHDhlq/fr3atm2r+vXra+7cuSpatGhBhwgAAADgKlfoEiljjCZMmKBJkybpwoULCg0NVe/evdW9e3f5+voWdHgAAAAAUHhu7XNyOBzq16+fNm/erJ07d2rNmjXq2bMnSRQsc+fO1Z133qmyZcuqWLFiqlKlinr37q2dO3cW6DL27Nkjh8OR7SsoKCjP4gSQO97anxhj9Ndff+njjz9Wnz59dMMNN8jf318Oh0PFihUrFDECyHt5/XtNTEzUN998oyeffFINGzZUaGioihQpotKlS6tVq1Z6//33lZiYWGDxXfUMcIVIS0szPXr0MJI8voKCgsz8+fMLbBm7d+/OdL6Mr8DAwFzFCCD3vL0/yWp/4O/vXyhiBJB3vPV7DQ4Ozva4ombNmmbbtm0FEt/VjkQKV4zhw4dbP/gHHnjA/Pnnn+bo0aNm3rx5JioqytoR/P333wWyjIwHTt9//705c+aMx9fZs2dzsxkA5AFv708y7g8qVqxo7r33XnPTTTfZSqTyY58HIG946/fq3Gc89NBDZubMmWbnzp3m+PHjZv369eapp54yDofDSDJRUVHmzJkz+R7f1Y5ECleEw4cPm8DAQCPJ3HXXXSYtLc2lfNeuXVb5/fffXyDLyHjg9Msvv1xWDAC8Lz/2J6dPnzZz5841//zzjzVtyJAhOU6k8iNGAHnDm7/XPn36mMOHD2daPmrUKOvYY/To0fke39WORApXhDfffNPaUaxfv95jnaefftpIMj4+PubIkSP5vgwSKeDKkB/7E0/sJFIFFSMA+wry95qcnGzCw8ONJNOiRYtCF9+/XaHrbAL5Z/Xq1Wrfvr0iIiJUtGhRBQQE6Prrr9eUKVMKOjQ3cXFxkqTq1aurXr16Huvcf//9kqS0tDR99913hXIZwL8V+5O8dSXECHgT+5Sc8fPzU/Xq1SVJhw4dKnTx/duRSF2FjDEaNGiQGjdurG+//Vb//POPkpOTlZiYqPXr16tXr156/vnnCzpMF+vWrZMkNW7cONM6MTExVu+Oa9euLRTLSEpKsh0HcCVhf+IdV0KMgDewT7HvyJEjkqSQkJBCGd+/GYnUVcYYo+7du+uNN95QiRIl9M4772jPnj06fvy4FixYoCpVqkiSxo4dqz///DPLtlJTU3X27NlcvXLi4MGDVl1nfJ74+/urfPnykqS///47R217axl9+/ZVUFCQ/P39VaxYMV1//fUaNGiQjh49aisuoDBjf+IdV0KMgDewT7Fv3bp12r17tySpSZMmhS6+fzu/gg4A+WvixImaMWOGSpUqpd9++001atSwyu68807FxcWpdu3aMsbok08+yfQSsCQtXbpULVu2zFU8JgfjQR87dsx6X7p06SzrlilTRgcOHFBCQoKtOPJ6GZs3b7beX7x4UevXr9f69ev13nvv6YsvvtAdd9xhKz6gMGJ/4h1XQoyAN7BPse+FF16QlD4Oa69evQpdfP92XJG6ihw/flyvvvqqJGnKlCkuOyina6+9VjExMZKU7dme/HLu3DnrfXYDWRYvXlyScnwmKS+X4ePjozvuuEMff/yxNm7cqJMnT+rcuXNau3at+vXrJ19fX504cUL33nsvl81xxWN/4j1XQoxAXmOfYt9bb72lRYsWSZKeeuop1alTp1DFdzUgkbqKTJkyRWfOnNH111+vDh06ZFqvUqVKkqTz589n2V6LFi1k0nt+vOzXv0mlSpX0ww8/6JFHHlHt2rVVokQJ6+HYd999V7Nnz5bD4VBiYqKeffbZgg4XyBX2JwDyEvsUe3788Ue98sorkqQ6depozJgxBRzR1YlE6iry5ZdfSpIefvjhLOs5d05hYWFejyknAgMDrfcXLlzIsm5iYqIkKSgoqNAto0OHDnrggQckSUuWLMm0dx3gSsD+xHuuhBiBvMY+JefWrFmjTp06KTU1VZGRkZo/f751NakwxHc14Rmpq8TJkye1fv16SdJNN92UZd1t27ZJktWdZmZSU1OtH93lysmPtVSpUtb7+Pj4LOs6y8PDw23FkR/LkKR27dpZ/yzWr1+viIgI220ABY39iXddCTECeYl9Ss5t27ZNbdq00ZkzZ1S6dGn997//VWRkZKGJ72pDInWVWLVqlXWZ2tN9x04HDx7Uzp07JUk333xzlm3m14OcERERCgoK0tmzZ62eaTy5ePGidZUnq3UsqGVI6Q9yOp08edL2/EBhwP7Eu66EGIG8xD4lZ/bv36/WrVsrPj5eISEh+uGHH1SzZs1CE9/ViFv7rhIZH8oMCAjItN5nn30mKf0szG233eb1uHLC4XDo+uuvlyStWLEi03pr1qxRamqqJKlBgwaFbhmSdPjwYet9aGio7fmBwoD9iXddCTECeYl9Svbi4+PVunVr7du3T8WLF1dcXFyO2mF/4l0kUleJjDup/fv3e6xz6tQpjR07VpL06KOPKjg4OMs28/NBznbt2klKv6S9ceNGj3Vmz54tKb33vLvvvjvHbefnMubNm2e9d+7YgCsN+xPvuxJiBPIK+5SsnT59WnfccYf+/vtvFSlSRLNnz872ilx+xndVM7gq1K5d20gyksyQIUPcyi9evGjuuusuI8mULFnSxMfH53+QWTh8+LAJCAgwkky7du3cynfv3m0CAwONJHP//fcXyDIOHDiQZftffvmlcTgcRpJp2bLlZcUIFAbsTy7PkCFDjCTj7+9faGMECgL7lMwlJiaam2++2UgyPj4+ZubMmYUqvqsdidRV4MKFC8bPz89IMpUqVTJFihQxY8aMMYcOHTIJCQlm/vz5JiYmxkgyfn5+5r///W9Bh+zR8OHDrR1t586dzYYNG8zRo0dNXFyciY6ONpJMUFCQ+fvvv93m/eWXX6x5P/roI68so1SpUqZTp07m008/NZs3bzbHjh0zR48eNUuXLjWPP/64lUQFBgaa9evX5+WmAfIN+5Oc7082b95sli9fbr169uxpJJmiRYu6TF++fLm5cOFCnsYIXCnYp2S+T0lJSTH33HOPVTZu3Dhz5syZTF/nzp3zSnzIHInUVWDt2rVGknE4HOa3334zQUFB1o8p4yskJMTMnz+/oMPNVFpamunRo4fH2J07gMziz+mBT26WUaJEiUznc74iIiLM4sWL82JzAAWC/UnO9ye33HJLtvsE52v37t15GiNwpWCfkvk+Zffu3Tneh0gylStX9kp8yBzPSF0FnF2KVqpUSc2bN9eyZct07733qkyZMvL391f16tX17LPP6u+//1bbtm0LNtgsOBwOTZs2TV9//bXuuOMOlS5dWv7+/oqOjtaTTz6p9evX5zr+3Czjww8/1NNPP62mTZuqUqVKCggIUNGiRVWuXDndcccdmjBhgrZu3apbbrklVzECBYn9CTECeYl9CvFdyRzGFPKhm5FrTz/9tN59913ddddd+u677wo6HABXMPYnAPIS+xRcybgidRVw9oZz3XXXFXAkAK507E8A5CX2KbiSkUhdBZw7qdq1axdwJACudOxPAOQl9im4kpFI/cvt27dPJ06ckMROCkDusD8BkJfYp+BKRyL1L+c80+Pr66tatWoVcDQArmTsTwDkJfYpuNKRSP3LOXdSVatWVbFixQo4GgBXMvYnAPIS+xRc6ei1DwAAAABs4ooUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAHkkLS2toENAIWCMKegQkA9IpIACtGPHDr388ssqV66cFi9eXNDh4Cryww8/qHXr1gUaw6lTp1S6dGklJiYWaBx27NixQy+99JLKlCnDb7aAPfnkkypZsqR+/vlnl+nnzp3Thx9+qCZNmujRRx/Nt3j279+v2NhYbd26Nd+WmZlz585p2rRpXtkG58+fV6NGjVS9enUdPnw4R/MkJibq448/VtOmTfP1MylIkydP1uTJk0ms/+VIpFConD9/XpMmTVKbNm0UERGhChUq6Prrr1fz5s01evRoHThwQEeOHFHdunULOtRcmzZtmp588kmNHj1aR44cydE8W7Zs0auvvqqYmBg5HA45HA75+/srKCjI5VW8eHGr3OFweHlNvOvgwYMaPHiwmjVr5rJODodDRYsWVWBgoCIjI9WqVStNmDBB58+fL+iQL9vFixc1depU1apVK0cH6adOndLLL7+satWqqUqVKrrlllvcDioz88UXX+jBBx90m75kyRK37exwOBQREaGkpCS3+mlpaZo0aZLq1q2rKlWqqG7duvrggw9ydDb266+/1p133qnixYtb0/bs2aPHHntM0dHRqlixoqpUqaKXXnpJp0+fzrKtRYsWqWXLloqOjla1atX0wgsv6Ny5cznYEjk3efJkPfbYY3rzzTcVHx+fq7aOHTum6OhoFSlSxG1bFylSROHh4apdu7Z69uyp33//PY/W4N/l008/1YkTJ/T1119b0zZv3qw+ffqof//+WrFiRb5dFVi4cKF69OihAQMG6Nprr1WPHj0UEhLi9tmGh4fr+eeft+ZbsWKFKleuLB8fH7fvQHR0tI4dO2Y7lrVr1+qJJ55Qv379vLINtmzZolWrVmnHjh367bffsq2/adMmPf7443ryySf1xx9/5CieSZMmqWzZsm7bz8fHR+XLl9eKFSvyYlW86oknnpAxRu3bt9eZM2cKOhx4iwEKiYULF5ry5cubihUrmvfff9+cPHnSKjt+/LgZP368KVOmjClevLipXLlywQWah1JTU01AQICRZH755Zccz3fu3DlTpEgRI8kMGzbMY53z58+b999/3zgcjjyKtmClpKSY8uXLG0mmY8eO5vjx48YYYw4cOGAmTJhgSpQoYSSZ6667zuzfvz/PlpucnGyWLl2aZ+15cuLECTNy5EhToUIFIylH34dDhw6Za665xjRt2tQcPXrUGGPMjBkzjK+vr5k6dWqW8yYmJprSpUtb2zCjW2+91Yoh42vkyJFudZOTk02HDh1MmTJlzJo1a4wxxmzYsMGULFnSdOvWzaSlpWUZx+23327mz59v/f3rr7+aoKAg8+STT5pz584ZY4z57bffTNmyZU2tWrXM4cOHPbbz1ltvGV9fXzNjxgxjTPr+olGjRqZevXoe1zGnfv75Z7dpKSkpl/WbzczBgwdNxYoVjSRTsWJFM3ToUPPuu++aZ555xvq+SzJPP/10ttvz3yir39+4ceNM06ZNzYYNG9zKnnzySSPJdOvWzcsRpv/vqlKlijl48KDL9OTkZNOrVy/rM3zjjTcy/Qw3bdpkgoKCjCRTqVIlc+jQoVzH9fjjj3tlG6SkpJhHHnnE3H333ebMmTM5nu+xxx6zHc/UqVOt7de4ceM82S75bdiwYaZRo0bmwoULBR0KvIBECoXC22+/bXx8fEzt2rXd/hlltGvXLhMZGfmvSaSMMdbBs92DMud8mSVSTm3bts1FdIVL8+bNjSTTs2dPt7Jly5YZh8NhJJlWrVrl2TI//PBDM2TIkDxrz5Njx46ZEydOmO3btxsfH59svw9paWnmpptuMn5+fmbXrl0uZffff78pUqSIWbduXabzz5kzx7Rr185t+rJly0zt2rXNX3/95fZKTEx0q//qq68aSeaTTz5xmT5x4kQjyUycODHTGI4ePWrKlCljkpKSjDHGHDlyxISHh5sGDRqY1NRUl7qLFi0yksytt97q1s5///tfj9+JjRs3Gknm/vvvzzSGrCQlJZlq1ap5LLvc32xmOnXq5PF7e+rUKes7L8l88MEHebK8K8nl/v6c301vJ1J79uwxISEhZvr06R7L58+fb31+f//9d5ZtNW3a1EgynTt3zpPY8msb5NSAAQNsx3PmzBlr+7333nveC86LUlNTzQ033GC6d+9e0KHAC7i1DwVu0aJFeu655+RwOPTpp58qIiIi07rR0dH69NNPr/jb1TLy8/Pz6nx9+/a9rPYLo6zWuVmzZmrevLmk9O/UgQMHcr085/Mw3hYeHq7Q0FBVq1ZNpUqVyrb+nDlztHTpUutWtoy6deum5ORk9e/fP9P5Z86cqYceesht+tChQzVw4EDVrFnT7VWsWDGXugcPHtSYMWMUFBSkTp06uZR17dpVPj4+euWVVzK9JW/WrFnq0KGDihQpIkkaP368EhIS1K5dO/n4uP5ruvXWW1W/fn39/PPPiouLcyl79tlnJUk9evRwmV67dm01bNhQs2fP1i+//JLptsjMiy++qB07dngsu9zfbGYCAgI8Tg8JCdEHH3xg/T1hwoQ8XW5hl5vfX15/Rpl59NFHVbJkSXXt2tVjecbPNrPP+dLyS39rlyu/tkFOOX/rdgQFBVnvQ0JC8jKcfOPj46PBgwdr+vTpmj17dkGHgzxGIoUClZiYqP/85z9KS0vTgw8+qHr16mU7z80336zbb789H6L7d2jTpk1Bh5BvIiMjrfe5TaR27typO+64I9fPwtiVk4Oojz/+WJKsxDGjJk2aSJJ+/fVX/fXXX27lZ8+e1S+//KJ77rnHZfrKlSu1aNEixcfH688//8w2hpkzZ+rixYuKiYmRv7+/S1mJEiV07bXX6syZM/rss888zn/pM1o//fSTJKlChQoe6zt/859++qk1bc2aNdq0aZOKFi2qG264wW2epk2bSpJLMpIdY4xef/11vfPOOzmex5tq166tkiVLSkr/Tl4tCur3Z8eCBQv0yy+/qH379vL19S3ocFCI3XnnnSpZsqQGDx5M5xP/MiRSKFAzZszQP//8I0nq0KFDjufr16+fy99JSUmaOHGiYmJidO211yo0NFR16tTRyJEjlZKSIin94e7y5ct77ITh2LFjKlOmjDU9KirKbZlz585Vo0aNVLVqVZcHxPfs2eNSb+PGjXrwwQdVp04dRUREqFy5crrvvvu0bt26HK9fXoiPj9drr73mNv3YsWMaMWKEKlasqMWLF2vr1q1q3LixwsLCNHPmTEnS6tWr9Z///Ec1atSQJE2ZMkVlypRR/fr1XXppOn/+vEaPHq0bbrhBNWrUUJkyZdSmTRstWLDAbblJSUmaMWOGGjZsqNdee01nzpzRgw8+qJCQED3zzDN5ss5btmyRlH7m0xn7pX777Tfdfffdqlu3rsqUKaOKFSvqkUce0a5du6w6v/76q+677z7rIG78+PGqVq2aqlWrpvXr11v1kpOT9c4776hp06aqVKmSQkND1a5du1x91tldbU1MTNSiRYskSVWrVnUrL1WqlHVV11kvo3nz5qlVq1YKDAx0mT5s2DClpKQoNjZW9evX17XXXqvPPvss0wfDnVeGPMUgyeoQxlMM+/fv1+7du3XLLbdY044fPy5JOnnypMf2nFfeMj5k7owhMjLS49luZww57YBDSr/C8NFHH1l/Oz/3S694OaWmpmrYsGGqWLGiSpQooe7du+dpL4TJyck6e/asJHm8Wv/bb7+pffv2uu666xQYGKi6devq//7v/9zq/fzzz2rfvr1atWolKf3qY1hYmFq2bOnSQcs///yj2NhY1apVS5GRkapatapiY2N19OhRtzaNMfr444918803q1q1agoKClLLli3drgAaYzRv3jw1adLE2ifFxcWpQYMGCgwM1C233OJy9S8nv78DBw5o2LBhio6O1vTp03O+Qf+/TZs26aGHHlLdunUVFBSka665RsOHD1dycnKO23jzzTclSW3btrW9/Jxau3atfH19Xf5vlSxZ0mU/9Ouvvyo4ONgqHz16tFs7xhi9++67qlatmooXL64mTZp43E/nZN+/du1a9enTR2FhYW7//5xmz55tfS/Kly+vLl26FFhS/Mcff6hTp06qV6+eypUrpxo1auill17SiRMnPNafMmWKrr/+ekVFRbls+8utJ6VfHWzevLn++usvrkr92xTojYW46t15553W/c/x8fGX3U7btm2Nr6+v9VByfHy8ueGGG4wkM3DgQJe677//vrXMSw0dOtRIcnsG69tvvzVly5Y1W7duNcak37f9xBNPGElm9+7dVr1ff/3V+Pv7m3vuucd6sHT69OlGkgkNDTXHjh1zW2blypUv63kL53yenpE6f/68efbZZ92eLRgxYoTLA+xz5swxNWvWtDquqF27tunQoYPx9fW1tsOsWbNMaGioNY+zI4NDhw6ZevXqmfbt25uEhARjTPozbM7t/uKLL7psvzp16lhtDBkyxLRr184EBwcbScbhcFjPymTllltuyfQZKed2lmSee+45j/N/8cUXxuFwmKeeesqkpKSYtLQ08/rrrxtJpkqVKm4xdOvWzYr3UufOnTM333yzee6556zPesaMGcbPz88UL17c/Prrr9mujyfZfR/Wrl1rreeSJUs81rnuuuuMJNOjRw+3srvuust88803LtMOHz5sWrZsaa677jpTtGhRl04m7rnnHo8PlIeEhBhJ5vXXX/cYQ58+fazteqm33nrL9O/f32VaixYtrI5EPHnvvfeMJBMQEGBNu+eee4wk07JlS4/zzJo1y1qPffv2eazjye7duzPdRxjzv89o4cKF5t577zUlSpQw5cqVs+aJjY3N8bKM+d/3zNOzfR9//LHV7muvveZSNn78eNOgQQOzfft2Y4wx27dvN9dee62RZPr06WOMMWbbtm3mtttus54fvOWWW8xbb71ldc4iyfz000/GGGP++OMPU6ZMGfPOO++Y5ORkY4yxfh+RkZHmyJEj1rJTU1PNAw88YB5++GFz6tQpY4wxP/zwgwkODjY+Pj5m5syZxhhjlixZ4tKByZAhQ8yYMWOMv7+/qVixohXXtdde6/ZsXGa/v5kzZ5o2bdpYbX700Udu223IkCGZPo8ze/ZsU6NGDbNq1SpjjDH//POPufnmm40k065duxx16nHgwAHrecZ//vkn03q//PKLFWd2neC0atXKY8x///239f2qUKGCx04LnPuFjNsi4zbo1q2bCQoKMpGRkVY8DofDfPjhh8aY9Oc0c7LvHzJkiLn++uutaRn//zk9/fTTxt/f33z55ZfGGGPOnj1revbsac1j95kt53yXPouZExMmTDAhISFm3rx5xpj0jjLee+894+PjYypWrGi2bdvmUn/8+PGmVq1aVqcWR48eNR06dHDbF+S0XkbOz+OBBx6wvR4ovEikUKDKli1rJJkiRYpcdhvr1q3zmPw4D0Cuv/56l+k7duzI9CDpp59+8thWq1at3A7wUlNTzXXXXefyj6R9+/Ye/7FHR0cbSdbOPKPcJlKlSpUyNWrUsF6VK1e2/sF7SgCOHz9uHSzfdNNNZvv27WbLli2ma9euJi4uzhjzv4PWUqVKmSeeeMKkpKSY6dOnmy5dulg9xN1+++0mNDTUOohyOnTokHWQfekD2M4kq3bt2mbOnDnm+PHjpn///mbUqFE5WudLE6mUlBSzbt0607dvX+Pr62uKFCliXnrpJZOSkuJx/nr16rlt64sXLxp/f38jyfz5558u9bNKpJ588kmPnR84e+mqXr2624FhTmT3fViwYIH1/b00XqdmzZpZSVBGx48fN6VLl86y96jExEQzc+ZMU61aNWs5d955p8vBZWJiolX27rvvemzH+WB5SEiIW1nDhg3N8uXLXaY5T3AULVrU7Nixw22eV155xUgyYWFh1rTGjRsbSebee+/1GIOzIwpJZu3atZmu86VymkjdeOONZurUqVbSMWjQIGud7fSw5/yeNW/e3Jw/f94Yk37wOWnSJKsntzZt2rh8bsuWLTPFihVz21a///67FXvGZP6FF16wEtsXX3zRpKSkmHHjxpnHHnvMJCYmWp1/9O3b16W9Q4cOWe2NGzfOmj5q1ChTs2ZNc/HiRZf6I0aMsE4cnT592ipv0qSJkWQaNmxonnvuOXPixAljzP86EpFkVq5c6XG7ZNbZRNWqVW0nUjt37jQBAQFuJyEOHDhgJXXO3h+z4vz/UqJEiSzr5UUiZYwxX331lZFkAgMDzdmzZ93KZ86caWrVquUyzbkNSpYsad566y3r+7Nq1SpTqVIlI8kUK1bMpYOnnOz79+/fn2kiNW3aNCPJDB8+3GV6cnKy9X8wvxKppUuXGh8fHzN06FC3MufvoVatWi6/q6pVq5pnn33Wpe6ZM2dMqVKlXKbltF5GGb8zzn0GrnwkUihQfn5+RpIpV67cZbdx6NAhExYWZjp16uQy3ZkURUdHu0zP6iDJ+U/v0kSqdu3aJjg42GzatMll+pgxY1z+kbz88sumWLFiVlfQTjfeeKORZD7++GO3ZXrjitThw4fNbbfdlukBSEREhJEy7wXJeQAaGBjochbaaeHChUbKvHepZ555xjqLnTGpefjhh40kt88qp5yJVHBwsKlWrZp1Rcv5jzu7rnEffPBBExoa6tYzpLP76UsPrjI7kDty5IgpUqSImTRpktsy/u///s+KyXnG247svg+ffvqp1b7zSsSlnD29tW7d2mX6lClTctxz1MWLF03Xrl2tZX3xxRdW2YEDB6zpU6ZM8Ti/s8ewS0+SbN++3e03aUz6gdbdd99tJJlrrrnGrFy50qSlpZljx46Z9957zzojn/HEiDPZ69Kli8cYnN9TSWbZsmU5Wm9jcp5IOc+4O2VMOpxXaXPC+T2rWLGiad++vWnSpImJiIgw0dHR5q677jKffPKJW1Levn1707RpU7e2Mia5zqtSxhgzefJk6zfpqQdGZ6K6fv16t7LXX3/d3HnnnVY348nJySY8PNy88sorbnUzJvqzZs2ypj/00ENGkundu7fbPFWqVDGSzFdffeVxu2S2H3OeMLCTSPXv399ERER4bM95Yu+uu+7yWH5pO5JMzZo1s6yXMZGqWrWqy0mvS1/ObvUzSzScJ4I89YbZpk0b884777hMc26Drl27utVfu3atlThmPJGV3b7fmPQeLT0lUs5hFXx9fa1EOaPY2Nh8TaSc+8G//vrLrSw+Pt46oZhxP+68anfgwAGX+peeYMhpvYyWLFlircvmzZttrQsKL56RQoFyDsR58eLFy26jfPnyio+P11dffSUp/bmd6dOn6/XXX5ekPHmws2XLljpz5owaNWqkgQMHKiEhQZL03HPPuTxPNXLkSJ09e1YNGjSQJG3YsEHPPPOMNm3alGex5ETZsmU1fPjwTMudz5Nce+21WZaXKlVKZcqUcSv/5JNPJMmtxzgnZ0cG+/fv19q1a3O83Jxq1qyZtm/frmXLllmdM6xdu1bly5fPcr4vvvhCx44ds541+f3339WrVy9r0Mucfj5LlixRcnKy3njjDbfe7UaPHq3w8HCFh4fr4MGDuVhLzy7t2MET58C54eHhLtMzG4TXk6JFi2r69OnW8x8ZO3nIbQydO3d2q+/n56e5c+dqwoQJCg0NVYcOHXTLLbdo4MCBql69utX734033pjjODIOIHxpHHnh0t9Gxh4XL2cw4Bo1amju3Llavny5Dh48qF27dum7776zekF0SktL06JFi7R582a371/9+vWt71/GwVydv70qVap47NDk22+/lSTVrFnTrWzw4MFasGCB6tSpI0lat26dEhIS9NFHH7ktv0+fPtbync+/Zlx+6dKl3dp3/m7tDqZ9Ob3ALVy4UCdOnPDYM6UxRuHh4Zk+O5ORszObjL3KZcf5TGpmL2fnKJl5+eWXJUlvvfWW9eyvJO3du1dLlizRI4884nE+Tx1hXH/99br55pslpXfa4pTdvj9jnUvNnz9f8fHxqlGjhkJDQ93KnZ2m5Ifdu3dr2bJlkjz/nypVqpS1vefNm2dNb9mypfbv36+6detqzJgx1u/40h4zc1ovo+DgYOt9xmeNcWUjkUKBcu7gTpw44XLQY5evr6+OHj2qF198Uffdd5/8/Pz0yiuv5FWYGjFihO69916dP39ew4cPV+XKlfXSSy957NrZ19dX3333ndq2bauPPvpIjz32WI56I8xrtWvX9lo38Zs3b5aUfrDtyXXXXWe937t3r1dikKQ6depo3LhxktK7BB8/fny28zi72b/tttv0ww8/aMCAASpbtqyt5TofsP7ggw/cDoZ2796tY8eO6dixY2rfvr3tdcpOxmTR2QnBpZwdNmQ8aD18+LC2bNlidTaQE76+vnr33XflcDhceowLCwuzkhg7MUiZd73uXF7fvn21YsUKHTx4UL/++qv+7//+T8eOHbMOsh9++GGrvnNbZBdDxjgaNGigUqVKub3y4rPKeIDpzZMmCQkJOnv2rO644w6PB+TO75+z85iccHa2kpO4nd//QYMGuS17586d1vJjY2OtebLaFzm76TaZdGySl/bs2aPatWt73G5HjhzRsWPHrAPwrDj3/c6TgfmhU6dOqlatmvbu3etyYmPKlCnq2LGjwsLCbLV3/fXXS8q8gxe7nB3BeEqW85vzf5SU/f+pjP+jJk+erBtvvFHHjx/XCy+8oMqVK2vUqFFuJ3tzWi+jjN3fe+q8BVcmEikUqJYtW1rvV65cedntzJkzR3Xq1NG1116rBQsWqGvXrnn6Dy4oKEhff/215s+frwYNGujcuXN68803VadOHZfepk6cOKH27dvrrbfe0rRp0/T222+7JBX5KSAgQEOGDPFK286zbxnPeGeU8R96iRIlvBKD01NPPaWOHTtKkl544YUsv0f79+/XTTfdpHnz5mnOnDkaOnSoxx4as+M8G/z3339fVsy5Ubt2bevqxKFDhzzWOXLkiCSpfv361rSvvvpK9957r+2xZapVq6YaNWq4nE319fVV7dq1bcewYcMGGWOs3vRyIi0tTWPGjJEk3XbbbVb37pKsExTZxRAZGWmdDT9+/LgSEhLcXqdOncpxTDnhzaTAG98/Z7w5abMgv/+5lZKSou3bt+c60XVeicrLHhqz4+vrqxdffFGSNHr0aKWlpSklJUUffvihnnjiCdvtOXvutJuAZcZ5Jc9Oz4fekvGKcHb/pzL+jypXrpx+/fVXffrpp6pWrZoSEhL0yiuvqHHjxi7t5LReRhmvKmeW3OHKQyKFAvX4449bZyrnzJmT4/lSU1OtndXPP/+szp07q2vXrurevXu28+bmKk3btm21Zs0aff755ypVqpT27dun3r17W+X33XeffvjhB3399dfZ3maWn06fPu1ym01uObu8znjbXkbOAy0fHx/rgNubpk6dqkqVKikpKUmdO3f2eGvOxYsXdfvtt2vXrl367LPPcpXgOcc6ymyMJCk9adu6detlLyMzJUqUsMaPct4ymtGRI0d0+vRpORwOlzHEsroSlJ3Q0FC3Mauct/x5ikGStm3b5lLvcmP4v//7P61du1YBAQF6//33Pcbw119/eTww9hTDnj17ZNKfD3Z5LV682FZcBalUqVIqWrSoNm7cqI0bN2Za77///W+O26xYsaIkZdk1s7Nbdef3f/bs2ZmegT979qyWL1+e4+XnlwoVKujkyZP6/vvvM62Tk+1Wrlw5SfZvR8ytbt26KSIiQlu3btXXX3+tb7/9VmFhYS63vOaUc8gB563oueU8WZFZl+j5Yd++fTp+/LjLsAzZ/Z+69I4Rh8OhLl266K+//tKECRMUGBioP//8UwMGDLisek5nzpyx3ufnbY7wLhIpFKjrrrvOOpP2wQcfWAc+2Xn99detg+X3339fqampqlSpkse6lx5gZby87vxH4uQ8K33pbYaPPfaYy98PPfSQFi5cKCl9HBcp/YDyl19+UWBgYKbPYxTUQHwDBw7M07OEzitAK1eu1P79+93KnbdKtGzZMl8SyrCwMH3++efy9fXVnj179Mgjj7hdEVi4cKG2bt2qcuXKZXo28NLPJ7Ok+5ZbbpGPj4/WrFmjYcOGuZUbY/Tcc8/ZvvqTU0899ZSk9Ge1LuU8eL311lutbb93714dOHDgsg62kpOTtXv3bpcTBpLUs2dPFSlSRMuXL3f7bh0/flzbtm1TWFiY7rrrLmv6zJkzc/yMlpT+LM6LL74oHx8fffbZZ6pevbpLeYsWLVSzZk2dPXvW5TkPJ+e26Nq1a46XKeXuZEt+KFKkiG688UYZY9S9e3ePV9O+++47LV26NMdtOm/5HD9+vMcD4dmzZ1u3gN1www0KCgrSoUOH1LdvX49X31577bVcJxne+Bycd0H06dPH4zOM69aty9EtkVWqVJGU/7doFS1aVM8++6yk9GdyP/jgA/Xq1euy2lq5cqWKFi3q8ZnFy+E82XLo0CFrTD9PvHm1dty4cQoKClKDBg2suw0yO0nr/D+V8XbhjP/r/fz81LdvX+uEmfN/vZ16GWX8nRbUnSrIeyRSKHDjxo3TrbfeqgsXLui+++7L9pmaESNGqH79+tZBlXOnPGXKFOsq1apVq6zb2k6ePKnk5GTrn2N4eLh1m5LzwdDU1FTNmDFDI0eOlJR+K8CpU6estlesWOF20Fq/fn2VLFnSOpPrrHvixAl98MEHktLPyo4ePdo6yDt27Ji2b9/uctDnTNrsJjrOs2nZJWc///yzNmzY4JJoOufJ7EAnu/KuXbvquuuuU1pamtWpR0ZxcXHy9fXVqFGjbLWbHed8nuZv3ry5Ndjnd999p+eff96l3Pn5bNiwwTobnZCQoJdeesm6Wnfs2DGtWLFCu3fvlvS/5x8uXLjg0lblypWthGDw4MF68MEHtWzZMh06dEi///677r//fqWlpalatWq219H5uaampmZap3PnzmrSpIkWLlzodlvb9OnT5efn5zIo58yZM/XAAw9kemCampqa6QP2Y8eO1YABA9ySmMqVK+vpp5/WyZMnNXfuXJeyGTNmKC0tTa+99pr1W1uxYoVKlizp1k5mVq5cqTvuuEPGGM2ePdvjgN0+Pj4aO3astd4ZrV+/Xn/++afuvfde2wlkxtuCL/3spf/9VrP6jOz8np23h2X2rJcnL730kqT0s+033HCDPvvsM+3bt09///23xowZo549e1oJt5T9b69///4qWrSozpw5o1atWunHH39UUlKS/vnnH40aNUpPP/20Hn/8cUnp28c5KPrUqVN1++23a+HChTp48KDWrl2rJ554QkuWLNGtt95qte/cjhk7SbjUpdsss9+fU1b7zszKnn32WRUpUkT79u1TTEyMPvjgA+3atUs7d+7UlClT1KZNG7cB3z25/fbbJaXvMzJeabhUxu2d3W2AWe3fMnriiSdUsmRJrV27VkuXLs20kwknT4nL0qVLtWbNGg0cONDl9uac7KMznmjMuH3btm1r7fOyuq3czu2QGf+/ZfXdkaRffvlFZ8+eVdGiReXj46M33nhDUvr+KOMt+M4YfvrpJ91zzz0u+4d58+a53bLq/B47/9fbqZfR9u3bJaUn4YXhOTLkkbzuBvD33383HTt29DgQZE6dPHnSvPTSS6Zq1aomOjra3HzzzWbRokV5GCUKm3PnzpnHH3/cOBwOU6pUKTN8+HCXbkVTU1PNokWLTLdu3cx3333nMq9z3Ar9/zExIiIizDXXXGO++OILa3qpUqVcuirOODhg6dKlTWhoqLnrrrvMjz/+aE2vVKmSefvtt40x6QOclipVysyaNcvqhnjGjBnG4XBY7SYlJVnjmkgyERERpnjx4mbw4MGmXbt2RpIpXry4ad68uTVuxV9//WV1Qfvyyy/neHudP3/eGvvIU9e2xqSPsTR79mwTGhpqJkyYYE3fvn27NQBv586dPY5n4RyYWJI1ttSltm3bZg3uO3z4cJOUlGTS0tLMvHnzTFhYmFvX4GfPnrUG5a1Ro4btAZjXrl1rihUrZnV/7mmsodTUVGtgV0nm/vvvN+vWrbO60Q4LC7PKKlasaAICAsykSZNM3bp1jZQ+/k+HDh2sMYCcXZnfeOONJjU11SxevNgavPTEiROmQYMGVnsZXzVq1PA4+HJ2duzYYX02o0ePzrLurl27TMWKFU2bNm3MmTNnTGpqqnnrrbeMn5+fW3fQ9erVy7Ir9ptvvtn4+PiYPn36WF0enz171owePdp88MEHmc538eJF07JlS1O5cmWzc+dOY0z6sANBQUHmsccecxlLqX///uatt97Kcp3S0tLM2rVrTd++fU2RIkVMo0aNctRN8IABA0zRokXN999/b4xJHyvouuuuMw0bNvTYDXN20tLSrO/2jz/+aC5evGiGDRtm0tLSzMaNG61x2i4dn2b16tXWd2DatGk5Wta+fftMhQoVrP2Ds4vxnHAOlnvpy8/Pz3z99ddWvdTUVGuf5+/vb1avXu2xvQ8//NBat4wvf39/s2DBApe6SUlJLgOqZ3yVLVvWpWv+c+fOWV3Vt2rVyqUr91OnTlnrf2k39ln9/g4fPmwNFvvggw+6zJeammoNOVGtWjW3AaU//PBDa+DZS1/jx4/P8fZ3Dm6b2W8rNTXVPPbYY1bbw4cPz3R8sU2bNlndn1eoUCHbAaSz6trc6aOPPjI+Pj7G39/fjB8/3vq/s3z5clO+fHnz/PPPu82Tk33/Dz/8YNW5dP+wZs0a63Pp3bu3te1/++03a+yqypUrmzVr1mQ6fENGGYcUeOqppzzWuXDhgvnwww9NUFCQtQ9wev75542UPl7Uli1bjDHp4+ndc889pmHDhm5dvAcGBpro6Gjre5aWlmaGDRtm/P39zdKlS23Xy+jpp582ksxLL72U7XrjypFnidTcuXNdRi+3O06A06FDh8w111xjmjZtag3+NmPGDOPr62umTp2aV+GikFq3bp159tlnTd26dU14eLgpWbKkad68ubnjjjvMpEmT3AZ/NSZ9BzZ48GBTrlw5ExISYh599FFz8uRJk5KSYpo3b27Kli3rNsDi2bNnzX/+8x8THBxsypcvb1599VWTnJxsfvnlFxMREWFGjhzpcgB23XXXWd/t4sWLm+rVq5tWrVq57TA3btxomjdvbooXL25q165tDcA7Z84cExISYtq2bWslEM6Dv0sPQC4dlyKjnTt3mvHjx1vjY2RM2qpWrWq9IiMjraTD19fXHD582BhjTPfu3d0OIEJCQsxvv/1mjEn/B1OqVCm3g4sGDRp4jCc+Pt7079/fREVFmfDwcFO7dm3TuXNnt4E1J0+ebIoXL+52cJbdgbUx6d8J5zhPGV++vr6mcuXKbtvr4MGDbutQsWJFY0z6OB7169c3xYoVM40aNbLGFnr33XdNcHCw6dq1qzUgqjHp/6C7dOliihcvbm666SYze/Zsl2WdO3fODBkyxFSrVs0UKVLEREREmL59+9oaQ8iY9O0eGRnp9tmULl06y8FBDxw4YLp3726io6NNlSpVTIcOHdwGnv3rr79MtWrVslz+Dz/8YGJiYkxgYKAJCQkxd955p3nttdfM3r17s439woULZsiQIaZ69eqmSpUqplmzZi4H8cakH1Bmd3A4cuRIExQUZCpWrGg6depk5s6da2tQ208//dRcf/31pkqVKua6664zY8aM8TheUk79+OOP1ve6T58+JiEhwfTu3dst0ShTpow5cOCAueWWW6wkWJJxOBzmxhtvzLT9+Ph4l8GzM84XGRlp/vjjjxzFGRcXZ26++WYTGBhogoKCTOvWrV3GzNqwYYN1YJtxGZkNYrxkyRLTqlUr67vQoUOHTAd+Tk5ONm+//bapXbu2KVq0qClVqpR55JFHXD7nL7/80hpU2PkKCwszs2fPNgMHDjSBgYEuZaVKlbKS8sx+f6NHj3bbnzjnW7dunSlZsqRLWWBgoNsYS8uWLTNt2rQxISEhpnjx4qZZs2aZJg6ZGT58eKbJ16OPPuq23lL64LjPPfecVe+PP/4wFStWtE6oZdy/RUVFZXrC6ejRo8bhcGR60O60ceNG89RTT5mqVauaEiVKmLp165r27dubxYsXu9TL6b6/ffv2bvup+vXru9TZvn27eeCBB0xoaKgJDQ01bdq0MWPHjjWvvPKKiYiIMF26dDHTpk3Lcty//fv3m7i4ONO6dWu3fWKVKlVM1apVTZUqVUxERIQVT8mSJU1SUpJbW998841p0aKFCQ0NNdWrVzcNGzY0Y8aMcdnXO2X8PoaEhJjq1aub9u3bu53gyGm9jGJiYoyPj4/ZunVrpnVw5XEYkzc3q+7cuVNRUVFq3bq1fvnlF3Xr1s3tVovsGGN0yy23aPny5dq2bZtL3/+dOnXSvHnztHLlSpdeoAAAAPLbuXPnVKVKFdWrV89Wxx55Yf78+XrxxRdduvlG4XXkyBGVL19eXbt21YwZMwo6HOShPHtGqmrVqvL19dUNN9xw2W3MmTNHS5cuVcuWLd0GUOvWrZuSk5PVv3//3IYKAACQK4GBgXrrrbf0888/W8+/5JdJkyZddicTyH9TpkxRyZIl9eabbxZ0KMhjed7ZhKcR03Pq448/liS3bnYlWWOH/Prrr/rrr78uexkAAAB54ZFHHlHv3r3z9STv5s2btWTJkmw7mUDhcOzYMb3zzjv66KOPrG7z8e+R54nU5XZXmpiYqEWLFkmSS///TqVKlVJERIQkWfUAAAAK0rhx4xQaGqqhQ4d6pf1ly5apUaNG6tKli6ZOnaqOHTuqX79+eTaQLrzHObbh2LFj1a5du4IOB15QaLo/37p1q9UlZmbjATl3GuvWrcu3uAAAADLj5+enzz//XP7+/nr99dfzfLzAmTNnatWqVfr888/1+OOPq3Tp0pkO+orC49ChQ+rRo4cGDRqkbt26FXQ48BLvjBZ5GY4cOWK9Dw0N9VinRIkSkmSNFZTR5MmTNXnyZEnpSVnNmjXzPkgAAIBMXLx4UbNmzcrVYw6XOn/+vPz9/ZWamqqSJUvq3Llzuummm/KsfXjH6dOnFRQU5DamIa48e/bs8Zh7SIUokUpISLDeBwQEeKzjvG3Q02BuvXr1sh68jImJ0erVq70QJQAAAICrRUxMTKZlhebWPn9//2zrOEfTDg8P93Y4AAAAAJCpQpNIlS9f3np/9uxZj3VOnjwpSSpdunR+hAQAAAAAHhWaRKp27dry8UkP59ChQx7rOJ+jYkBeAAAAAAWp0CRSJUqUsMaP2rRpk1v5kSNHdPr0aTkcDrVp0ya/wwMAAAAAS6FJpCTpqaeekiQtWbLErWz58uWSpFtvvdXlNkAAAAAAyG95nkilpKRIklJTUzOt8+qrr6px48b67bffXKZ37txZTZo00cKFC91u75s+fbr8/Pw0evTovA4ZAAAAAGzJ00QqNTVVK1eulCRt3rxZFy9edKuTkJCgESNGaOXKlW5JkY+Pjz7//HOVLVtWjz32mM6ePau0tDSNGTNG8+fP15QpU9SwYcO8DBkAAAAAbMuzRKpnz54KCwvTwoULJUnr1q1TyZIldc8997jUK1mypDp27KiQkBB16dLFrZ3o6Gj98ccfKlu2rOrWravq1atr2bJlWrlypbp3755X4QIAAADAZXMYY0xBB5HXGJAXAAAAQG5llVcUqs4mAAAAAOBKQCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIFVJ9+/ZV3759CzoMAP8C7E8A5BX2J8D/kEgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANiUZ4nUrFmz1LhxY1WpUkU1a9bUiBEjlJycbLsdY4w+/PBDNW3aVGXKlFFkZKRq1aqlwYMH6/z583kVLgAAAABctjxJpGJjY9WtWze98cYb2rVrlxYuXKhp06bprrvu0sWLF2219cgjj6hPnz6KjY3VkSNHtH//fs2aNUuzZs1Ss2bNdPr06bwIGQAAAAAuW64TqalTp2rixIkaMGCAWrduLUmKjIzU+PHjtXDhQg0cODDHbc2bN0+ffvqpBg4cqIcfflgOh0OSVLt2bb3//vv6888/NXr06NyGDAAAAAC5kqtE6vz58xowYIAkqUePHi5lbdu2ValSpTRu3Djt2LEjR+0tXLhQktSgQQO3skaNGkmS1q1bl5uQAQAAACDXcpVILViwQPHx8YqKilJERIRLmcPhUJMmTZSWlqYpU6bkqL3AwEBJ0m+//eZW5rylr2HDhrkJGQAAAAByLVeJVFxcnCSpatWqHsvr1q0rSVq0aFGO2uvYsaMk6e2333a78vTVV18pKipKzz777OWGCwAAAAB5IleJlDPZqVSpksfysLAwSdKGDRuUlpaWbXuNGzfWM888o8TERN1666368ccfJUmrV6/W3Llz9euvv1ptAgAAAEBB8cvNzEeOHJEkhYaGeiwvUaKEJCk5OVmnTp3KURI0duxY+fj4aOzYsbrrrrvUr18/paam6r///a+KFCmSm3ABAAAAIE/kKpFKSEiQJAUEBHgsd/a6J0mJiYk5SqQcDofeeustHTx4UGXLltU777wjf39/XX/99erevXum802ePFmTJ0+WJMXHx9tYCwAAAACwJ1e39vn7+2dZnpSUZL0PDw/PUZvnzp1T+/bt1bt3b73zzjv68ssvZYzRo48+qjfeeCPT+Xr16qXVq1dr9erVKl26dM5WAAAAAAAuQ64SqfLly0uSzp4967H85MmTkqTg4OBsky5JSktLU4cOHeTv76+bbrpJktSpUyfFxcXJ399fgwYN0ty5c3MTMgAAAADkWq4SqXr16kmSDh065LHc+QxV/fr1c9Te559/rp9++knt2rVzmd66dWtNmzZNkqxxqwAAAACgoOQqkWrbtq0kadOmTR7Lt23b5lIvO/PmzZMklS1b1q2sS5cuqlevnv766y+dOnXqcsIFAAAAgDyRq0SqU6dOCg8P19atW62rT06pqalatWqV/Pz81Llz5xy1l5KSIknat2+fx/Lq1avLx8dHvr6+uQkbAAAAAHIlV4lUcHCwhg4dKmOMZsyY4VIWFxenhIQExcbGKjo62qXs1VdfVePGjfXbb7+5TL/vvvskpd/id6mkpCStWbNGd9xxh4KCgnITNgAAAADkSq4SKUnq3bu3unTpohEjRmjNmjWSpPXr16t3795q06aNRo0a5VI/ISFBI0aM0MqVKzV69GiXsi5duqhr165avHixYmNjrU4sEhIS1K1bN6WlpemDDz7IbcgAAAAAkCu5GkfK6ZNPPtGECRP0yCOP6MKFCwoNDdWwYcPUvXt3t9vwSpYsqY4dO+qnn35Sly5dXMocDodmzJihFi1aaNq0aapQoYLCwsJUrFgx3XPPPXrvvfdUsmTJvAgZAAAAAC5bniRSDodD/fr1U79+/XJUd86cOVmW9+zZUz179syL0AAAAAAgz+X61j4AAAAAuNqQSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgU56MI4WsRVWurL379tmap3bt2pLSx9Wyq3KlStqzd6/t+QAUflGVIrV3/wFb8+RqfxJZUXv27bc9H4DCj/0JkDskUvlg7759Sjv4t615Yl8bLUna8GPmgxdnxqdCDdvzALgy7N1/QCkLP7Y1T/+P4iRJ699+wfby/Fp3sz0PgCsD+xMgd7i1DwAAAABsIpECAAAAAJtIpAAAAADAJhIpAAAAALCJRAoAkKf69u2rvn37FnQYAAB4FYkUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNfgUdgDecOXNGr7/+uvX3448/LkmaMmWKNe2WW25RixYtNHbsWJ09e1aSVL58efXq1UtxcXFau3atVffZZ5/VoUOHNHPmTGva3XffrYYNG7os55prrtFDDz2kL774Qtu2bfPa+uWEMy5vrdOQIUO0Zs0afffdd9a0Bx98UBERERo3bpw1rUGDBmrXrp0mT56sf/75R5IUFBSk5557TosXL9aSJUusugX9ObFOrNOVsE4FwbkOOV2nvXv3qnLlylf158Q6sU5XyjrlN+f68jmxTlfKOmXFYYwxWda4AsXExGj16tUFHYbF4XAo7eDftuaJfW20JGnCay/ZXp5PhRr6F36sAJS+P0lZ+LGtefp/FCdJevfRrP8heOLXupvt/Unfvn0lSRMnTrS9PAD5h/0JkL2s8gpu7QMAAAAAm0ikAAAAAMAmEikAAAAAsIlECgAAAABsIpECAAAAAJtIpAAAAADAJhIpAAAAALCJRAoAAAAAbPLLq4ZmzZqlMWPGKD4+XkWLFtUjjzyiF154QUWKFMlVu2fOnNGnn36qRYsWKTg4WBEREerVq5cqV66cR5EDAAAAgD15kkjFxsZq2rRpmjdvnlq3bq39+/erRYsWWrx4seLi4uTv739Z7X711VeKjY3Vvffeq//7v/9T6dKl8yJcAAAAAMiVXN/aN3XqVE2cOFEDBgxQ69atJUmRkZEaP368Fi5cqIEDB15Wuy+99JK6dOmit99+Wx988AFJFAAAAIBCI1eJ1Pnz5zVgwABJUo8ePVzK2rZtq1KlSmncuHHasWOHrXYHDRqkN998U5MmTdLDDz+cmxABAAAAIM/lKpFasGCB4uPjFRUVpYiICJcyh8OhJk2aKC0tTVOmTMlxmzNnztQbb7yhe++91y05AwAAAIDCIFeJVFxcnCSpatWqHsvr1q0rSVq0aFGO2jt9+rSeeeYZORwODR8+PDehAQAAAIDX5CqRWrdunSSpUqVKHsvDwsIkSRs2bFBaWlq27U2aNEmHDx9W48aNtWPHDnXp0kWNGjVSZGSk2rVrpxUrVuQmXAAAAADIE7lKpI4cOSJJCg0N9VheokQJSVJycrJOnTqVbXtffPGFJCk+Pl5nzpzRlClTtHz5cg0ePFg//PCDbrrpJn3zzTce5508ebJiYmIUExOj+Pj4y1gbAAAAAMiZXCVSCQkJkqSAgACP5Q6Hw3qfmJiYZVunT5+2rnBNnDhRDz/8sAICAuTr66vHH39czz33nJKTk9WtWzdruRn16tVLq1ev1urVq+nhDwAAAIBX5SqRym58qKSkJOt9eHh4lnX3799vva9QoYJbee/eveXj46MzZ87oq6++shkpAAAAAOSdXCVS5cuXlySdPXvWY/nJkyclScHBwdkmXadPn7beh4SEuJVXqlRJderUkSRt2rTpcsIFAAAAgDyRq0SqXr16kqRDhw55LHc+Q1W/fv1s2ypTpoz1PmNSlVHFihUlScYYO2ECAAAAQJ7KVSLVtm1bSZlfIdq2bZtLvaxER0dbzzZt2LDBY53ixYtLkmrUqGE7VgAAAADIK7lKpDp16qTw8HBt3brVuvrklJqaqlWrVsnPz0+dO3fOPhAfHz344IOSpPnz53uss2vXLvn5+aljx465CRse9O3bV3379i3oMAAAAIArQq4SqeDgYA0dOlTGGM2YMcOlLC4uTgkJCYqNjVV0dLRL2auvvqrGjRvrt99+c5k+cOBAlSxZUrNnz9bOnTtdytavX69169apX79+ioyMzE3YAAAAAJAruUqkpPTe9Lp06aIRI0ZozZo1ktKTnt69e6tNmzYaNWqUS/2EhASNGDFCK1eu1OjRo13KypQpo3nz5snPz08dOnTQnj17JKVfierWrZvH9gAAAAAgv/nlRSOffPKJJkyYoEceeUQXLlxQaGiohg0bpu7du8vX19elbsmSJdWxY0f99NNP6tKli1tbN954o1asWKFBgwapQYMGKlWqlEqUKKHHH3/c6gIdAAAAAApSniRSDodD/fr1U79+/XJUd86cOVnWqVOnjubOnZsXoQEAAABAnuPyDgAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGATiRQAAAAA2EQiBQAAAAA2kUgBAAAAgE0kUgAAAABgE4kUAAAAANhEIgUAAAAANpFIAQAAAIBNJFIAAAAAYBOJFAAAAADYRCIFAAAAADaRSAEAAACATSRSAAAAAGCTX0EHAM8mvPZSQYcA4F/i3UfbFXQIAP4l2J8A/8MVKQAAAACwiUQKAAAAAGwikQIAAAAAm0ikAAAAAMAmEikAAAAAsIle+wAAmfIvWkQOh8PWPLVr15Yk2/NVjozUnn37bM0DAEBBIZECAGTqYlKyUpd/a2uefu9/Lkn6c8oIW/P5Nr3HVn0AAAoSt/YBAAAAgE0kUgAAAABgE4kUAAAAANiUZ4nUrFmz1LhxY1WpUkU1a9bUiBEjlJycnCdtT5o0SQ6HQ9OnT8+T9gAAAAAgN/Kks4nY2FhNmzZN8+bNU+vWrbV//361aNFCixcvVlxcnPz9/S+77b///lvPPvtsXoQJAAAAAHki11ekpk6dqokTJ2rAgAFq3bq1JCkyMlLjx4/XwoULNXDgwMtuOzk5WV26dJGPD3cgAgAAACg8cpWhnD9/XgMGDJAk9ejRw6Wsbdu2KlWqlMaNG6cdO3ZcVvsDBw5UnTp11LBhw9yECQAAAAB5KleJ1IIFCxQfH6+oqChFRES4lDkcDjVp0kRpaWmaMmWK7bYXL16sefPmacKECbkJEQAAAADyXK4Sqbi4OElS1apVPZbXrVtXkrRo0SJb7Z44cUI9e/bUp59+qqCgoNyECAAAAAB5LleJ1Lp16yRJlSpV8lgeFhYmSdqwYYPS0tJy3O4TTzyhXr16KSYmJjfhAQAAAIBX5CqROnLkiCQpNDTUY3mJEiUkpXcacerUqRy1+dFHHykhIUEvvPCCrVgmT56smJgYxcTEKD4+3ta8AAAAAGBHrro/T0hIkCQFBAR4LHc4HNb7xMRE6wpVZnbu3KmhQ4dq2bJltnvq69Wrl3r16iVJV/2VLH//oi7bPidq164tSbbnq1y5kvbs2WtrHgAAAOBKl6tEyt/fXykpKZmWJyUlWe/Dw8OzbCslJUVdunTRO++849ZxBey5eDFJaUf32JonduAbkqQNP39naz6fMlG26gMAAAD/Brm6ta98+fKSpLNnz3osP3nypCQpODg420F5X3/9dV1//fVq3759bkICAAAAAK/LVSJVr149SdKhQ4c8ljufoapfv362bY0cOVIffPCBHA6H22vJkiWSpEcffVQOh0MtWrTITdgAAAAAkCu5urWvbdu2mjNnjjZt2uSxfNu2bVa97FSrVi3Tsn379ikxMVHlypVTiRIlMu0lEAAAALgcffv2lSRNnDixgCP5f+3dfVBU593/8c8CsooQUSCJpCBondgWnyoJpjaKaanxsdFRsbWxqCMTDVCbu83Uh9x2NKHaKJNE299EYlK1/SWpmmgIURNttNKYxqdV40RtENRERcSgoIAI5/7DYW+5BdzDWViWfb9mdmaH65zr+saZ/YYP5+x14C0sBalJkybp2Wef1fHjx1VUVKT77rvPOVZTU6N9+/YpICBAycnJd53r+PHjjY4lJiZq9+7d+sMf/qCUlBQrJQMAAACAZZZu7QsJCdHixYtlGIbWrVtXbywnJ0clJSVKT09XbGxsvbEFCxYoISFBeXl5VpYHAAAAAI+wFKQkac6cOZo6daoyMzN14MABSZLD4dCcOXM0cuRILV26tN7xJSUlyszM1GeffaZly5ZZXR4AAAAAWp2lW/vqrF+/XitXrtS0adNUWVmp0NBQLVmyRCkpKfL39693bLdu3TRhwgTt2LFDU6dOdcfyAAAAaOPsgR1a7TmXktQjKkqFZ86YPg9wlVuClM1mU0ZGhjIyMlw6dtOmTabm37VrVzMrAwAAQFtQdaNaNXvfM3VOxp//vyTpcHam6fX8Hxln+hzADMu39gEAAACAryFIAQAAAIBJBCkAAAAAMIkgBQAAAAAmEaQAAAAAwCSCFAAAAACYRJACAAAAAJMIUgAAAABgEkEKAAAAAEwiSAEAAACASQQpAAAAADCJIAUAAADAbdLS0pSWlubpMlocQQoAAAAATCJIAQAAAIBJBCkAAAAAMIkgBQAAAAAmEaQAAF7LV77QDABW0S/djyAFAAAAACYRpAAAAADAJIIUAAAAAJhEkAIAAAAAkwhSAAAAAGASQQoAAAAATCJIAQAAAIBJBCkAAAAAMIkgBQAAAAAmEaQAAAAAwCSCFAAAAACYRJACAABKS0tTWlqap8sAAK9BkAIAAABaGX+88H4EKQAAAAAwiSAFAAAAACYRpAAAAADAJIIUAAAAAJhEkAIAAAAAkwhSAAAAAGASQQoAAAAATCJIAQAAAIBJbgtSGzZsUEJCgnr27Kk+ffooMzNT1dXVpucpLi5Wenq6oqKiFBgYqAceeEAzZ85UYWGhu0oFAAAAAEsC3DFJenq61qxZoy1btigpKUlnz55VYmKidu3apZycHNntdpfmOXfunH7wgx/o9OnT6tChg2pqanTu3Dm9/vrreuedd7R9+3Y9/PDD7igZAAAAAJrN8hWp1157TatWrdL8+fOVlJQkSYqKitIrr7yijz76SAsXLnR5rrS0NMXExOjTTz9VRUWFysvL9eqrr6pTp04qLS3VlClTmnWVC21LWlqa0tLSPF0GAAAA0GyWgtT169c1f/58SdKMGTPqjY0aNUrh4eHKysrSl19+ede5zp8/r0uXLmn79u1KSEiQv7+/OnXqpNTUVL300kuSpIKCAuXl5VkpGQAAAAAssxSktm7dquLiYsXExCgyMrLemM1m0+DBg1VbW6vs7Oy7zlVYWKjly5c3eBvgrFmz1LVrV0nSxYsXrZQMAAAAAJZZClI5OTmSpF69ejU43q9fP0nSzp077zrXI4880uj3n2w2m3r37i1JiomJaUalAAAAAOA+loLUoUOHJEnR0dENjtddRTpy5Ihqa2utLKXi4mJFRUXpoYcesjQPAAAAAFhlKUgVFRVJkkJDQxsc79KliySpurpaV65cafY6//nPf1RQUKBnn31Wfn4Nl7x69WrFx8crPj5excXFzV4LAAAAAO7GUpAqKSmRJAUFBTU4brPZnO8rKiqavU5WVpYGDRqk2bNnN3pMamqq9u/fr/379ysiIqLZawEAAADA3Vh6jpTdbtfNmzcbHb9x44bzfVhYWLPWOHjwoDZs2KC9e/fK39+/WXMAAAAAgDtZuiLVvXt3SVJ5eXmD46WlpZKkkJAQlx/Ke7urV69q+vTpevPNN52bTQAAAACAp1kKUv3795cknTt3rsHxuu9QDRgwwPTcN2/e1JNPPql58+Y5H/QLAAAAAG2BpSA1atQoSdLnn3/e4PjJkyfrHecqwzA0a9YsjRs3TlOmTLFSIgAAAAC4naUgNWnSJIWFhen48ePOq091ampqtG/fPgUEBCg5OdnlOWtra5WamqpBgwZp5syZd4xfu3at0StgAAAAANAaLAWpkJAQLV68WIZhaN26dfXGcnJyVFJSovT0dMXGxtYbW7BggRISEpSXl1fv5zdv3tT06dPVt29fpaWl3bFefn6+JkyYoLKyMitlAwAAAIAllnbtk6Q5c+bok08+UWZmph577DENGjRIDodDc+bM0ciRI7V06dJ6x5eUlCgzM1OStGzZMv3whz+UdGt79MmTJ+uDDz5Qbm6uFi9e7DzHMAxVVVXp2rVrevjhh/Xggw9aLRsA0EJemfNzT5fQbtT9UXHVqlUergQA8H9ZDlKStH79eq1cuVLTpk1TZWWlQkNDtWTJEqWkpNyxZXm3bt00YcIE7dixQ1OnTnX+fPz48dq+fbuk/30+VUOmTZvmjpIBAAAAoNncEqRsNpsyMjKUkZHh0rGbNm264+fbtm1zRykAAAAA0OIsfUcKAAAAAHwRQQoAAAAATCJIAQAAAIBJbvmOFAAAANCW2AM7yGazuXx8XFycJJk6p06P6CgVnj5j+jx4N4IUAAAA2iQrj1OoulGtmqN7XD4+Y8WrkqTDb/4/02v5933U9DnwftzaBwAAAAAmEaQAAAAAwCSCFAAAAACYRJACAAAAAJPYbAIA0CaY3WFLYpctAL4pJqaHTpvsX83tl0FBQbp+/XqrrCVJPXr0UGFhoenzPIEgBQBoE8zusCWxyxYA33T69BnVXvra1DnpC34vSTqya7up8/zCH5BRaS5IpT3zX5Kko/s/M3WeJNk6Bpk+x1O4tQ9wo7S0NKWlpXm6DAAAALQwghQAAAAAmMStfZAkrXx+oadLAAAAALwGV6QAAAAAwCSCFAAAAACYRJACAAAAAJMIUgAAAABgEkEK7R5bkgMAAMDdCFIAAAAAYBJBCgAAAABM4jlSsMRut8tms5k6Jy4uTpJMnydJPXpEq7DwtOnz2qu6WxZXrVrl4UoA38BnDkBD7IGBrfr7ENoGghQsqaqqUu3l86bOSZ/335KkI//8yPR6ft26mz4HAACgJVXduKHaws9NnZP+/ApJ0pH33zK9nl9MnOlz4H7c2gcAAAAAJhGkAAAAAMAkghQAAAAAmESQAgAAAACT2GwCAIB2JiYmRqdPm9vh1NqOqj1UWFho+jwA8GYEKQAA2pnTp0/LqLxu6py0Z/5LknR0/2em17N1DDJ9DgB4O27tAwAAAACTCFIAAAAAYBJBCgAAAABMIkgBAAAAgEkEKQAAAAAwiV37AAA+yR4YaHqrb0tbhEdHq9DkluQA4C4rX/h9q621KmtFq63lSQQpAIBPqrpxQ7WFn5s6J/35W78cHHn/LdPr+cXEmT4HANB2cWsfAAAAAJhEkAIAAAAAkwhSAAAAAGCS24LUhg0blJCQoJ49e6pPnz7KzMxUdXV1s+a6cuWKfve73+nb3/62evbsqWHDhukf//iHu0oFAAAAAEvcstlEenq61qxZoy1btigpKUlnz55VYmKidu3apZycHNntdpfnOn/+vBITExUWFqa9e/cqIiJC69ev109+8hO9+uqrmjlzpjtKBgCgVdnt9lbdJRAA0LIsB6nXXntNq1at0pIlS5SUlCRJioqK0iuvvKIxY8Zo4cKFevHFF12ayzAMJScn69SpU9q2bZsiIiIkSU8++aTee+89zZ49W4MGDdKAAQOslg0AQKuqqqpS7aWvTZ2TvuD3kqQju7abOs8v/AFTxwMAzLN0a9/169c1f/58SdKMGTPqjY0aNUrh4eHKysrSl19+6dJ8mzZt0p49ezR8+HDFxsbWG/vlL3+p6upq/epXv7JSMgAAAABYZilIbd26VcXFxYqJiVFkZGS9MZvNpsGDB6u2tlbZ2dkuzbd27VpJ0pAhQ+4YGzx4sCTpn//8p7744gsrZQMAAACAJZaCVE5OjiSpV69eDY7369dPkrRz5867zlVRUeE8rqH5wsPDnWHNlfkAAAAAoKVYClKHDh2SJEVHRzc43rVrV0nSkSNHVFtb2+Rcx48fV0VFhUvz1a0LAAAAAJ5gMwzDaO7J999/v4qKivTrX/9aWVlZd4xnZ2crNTVVknT58mVnEGrItm3bNHLkSEnS4cOHnVezbjdkyBB98sknGjdunLZs2VJvbPXq1Vq9erWkW6GsT58+zf3PQhtWXFzs3IQEAKygnwBwF/pJ+1VYWKhLly41OGZp176SkhJJUlBQUIPjt2/XWlFR0WSQqpvLlfnqrlzdLjU11Rna0H7Fx8dr//79ni4DQDtAPwHgLvQT32Tp1r67PR/qxo0bzvdhYWGW5rp9vrvNBQAAAAAtyVKQ6t69uySpvLy8wfHS0lJJUkhIyF2DUt1crszHpVMAAAAAnmQpSPXv31+SdO7cuQbHi4qKJMmlB+jGxcXJz8/PbfOhfeL2TQDuQj8B4C70E99kKUiNGjVKkvT55583OH7y5Ml6xzWlS5cuzudHNTRfUVGRrl69KpvN5tyUAr6HRgXAXegnANyFfuKbLAWpSZMmKSwsTMePH3deLapTU1Ojffv2KSAgQMnJyS7NN3v2bEnS7t277xjbu3evJOmxxx6rdxsgAAAAALQ2S0EqJCREixcvlmEYWrduXb2xnJwclZSUKD09XbGxsfXGFixYoISEBOXl5dX7eXJysgYPHqyPPvrojtv7/vKXvyggIEDLli2zUjIAAAAAWGbpOVJ1fvGLXyg3N1c7duzQoEGD5HA4NGrUKA0YMECbN29WYGCg89iSkhKFh4dLksaMGaOcnJx6cxUUFGjo0KHq27ev/v73vysoKEhZWVmaN2+esrOzlZKSYrVcAAAAALDE0nOk6qxfv14rV67UtGnTVFlZqdDQUC1ZskQpKSny9/evd2y3bt00YcIE7dixQ1OnTr1jrtjYWH366adauHCh+vXrJ5vNpn79+umzzz7TwIED3VEuAAAAAFhi6da+OjabTRkZGTp27Jjy8/N14MABzZw5844QVXfspk2bdOXKFU2ZMqXB+R544AG98cYbOnXqlPLz8/Xuu+8SouC0efNmPf7447rvvvvUsWNH9ezZU3PmzFF+fr5H1ygsLJTNZrvrKzg42G11ArCmpfqJYRj64osvtHbtWj399NN66KGHZLfbZbPZ1LFjxzZRIwD3c/fntaKiQu+++66eeuopDRo0SKGhoerQoYMiIiL0ox/9SH/+859VUVHhsfp8ngF4idraWmPGjBmGpAZfwcHBRm5ursfWKCgoaPS821+dO3e2VCMA61q6nzTVD+x2e5uoEYD7tNTnNSQk5K6/V/Tp08c4efKkR+rzdQQpeI0XXnjB+YGfPHmycfjwYePixYvGli1bjJiYGGcjOHHihEfWuP0Xpw8++MAoKytr8FVeXm7lnwGAG7R0P7m9H3zrW98yxo8fbzz66KOmglRr9DwA7tFSn9e6nvGzn/3MeOutt4z8/Hzj8uXLhsPhMGbPnm3YbDZDkhETE2OUlZW1en2+jiAFr3DhwgWjc+fOhiRj9OjRRm1tbb3xU6dOOccnTpzokTVu/8Xp448/blYNAFpea/STq1evGps3bzbOnz/v/NmiRYtcDlKtUSMA92jJz+vTTz9tXLhwodHxpUuXOn/3WLZsWavX5+sIUvAKf/zjH52NwuFwNHjM3LlzDUmGn5+fUVRU1OprEKQA79Aa/aQhZoKUp2oEYJ4nP6/V1dVGWFiYIclITExsc/W1d27ZbALeaf/+/frpT3+qyMhIBQYGKigoSAMHDlR2dranS7tD3Tb5vXv3Vv/+/Rs8ZuLEiZKk2tpavf/++21yDaC9op+4lzfUCLQkeoprAgIC1Lt3b0m64xmsbaG+9o4g5YMMw9Bzzz2nhIQEvffeezp//ryqq6tVUVEhh8Oh1NRU/eY3v/F0mfUcOnRIkpSQkNDoMfHx8c6dIg8ePNgm1rhx44bpOgBvQj9pGd5QI9AS6CnmFRUVSZLuueeeNllfe0aQ8jGGYSglJUXPP/+8unTpopdeekmFhYW6fPmytm7dqp49e0qSVqxYocOHDzc5V01NjcrLyy29XPH11187j62rryF2u13du3eXJJ04ccKluVtqjbS0NAUHB8tut6tjx44aOHCgnnvuOV28eNFUXUBbRj9pGd5QI9AS6CnmHTp0SAUFBZKkwYMHt7n62ju3PJAX3mPVqlVat26dwsPDlZeXpwcffNA59vjjjysnJ0dxcXEyDEPr169v9BKwJO3Zs0fDhw+3VI9hGHc95tKlS873ERERTR5777336quvvlJJSYmpOty9xrFjx5zvq6qq5HA45HA49Kc//UlvvvmmRowYYao+oC2in7QMb6gRaAn0FPN++9vfSrr1nNbU1NQ2V197xxUpH3L58mUtWLBAkpSdnV2vQdX57ne/q/j4eEm66197Wsu1a9ec7+/2IMtOnTpJkst/SXLnGn5+fhoxYoTWrl2ro0ePqrS0VNeuXdPBgweVkZEhf39/ffPNNxo/fjyXzeH16CctxxtqBNyNnmLeiy++qJ07d0qSZs+erb59+7ap+nwBQcqHZGdnq6ysTAMHDtQTTzzR6HHR0dGSpOvXrzc5X2JiooxbOz82+9WeREdHa9u2bZo2bZri4uLUpUsX55djX375ZW3cuFE2m00VFRV65plnPF0uYAn9BIA70VPM2b59u+bNmydJ6tu3r5YvX+7hinwTQcqHvP3225Kkn//8500eV9ecunbt2uI1uaJz587O95WVlU0eW1FRIUkKDg5uc2s88cQTmjx5siRp9+7dje6uA3gD+knL8YYaAXejp7juwIEDmjRpkmpqahQVFaXc3Fzn1aS2UJ8v4TtSPqK0tFQOh0OS9OijjzZ57MmTJyXJuZ1mY2pqapwfuuZy5cMaHh7ufF9cXNzksXXjYWFhpupojTUkaezYsc7/WTgcDkVGRpqeA/A0+knL8oYaAXeip7ju5MmTGjlypMrKyhQREaEPP/xQUVFRbaY+X0OQ8hH79u1zXqZu6L7jOl9//bXy8/MlSUOHDm1yztb6ImdkZKSCg4NVXl7u3JmmIVVVVc6rPE39N3pqDenWFznrlJaWmj4faAvoJy3LG2oE3Ime4pqzZ88qKSlJxcXFuueee7Rt2zb16dOnzdTni7i1z0fc/qXMoKCgRo/729/+JunWX2F+/OMft3hdrrDZbBo4cKAk6d///nejxx04cEA1NTWSpO9///ttbg1JunDhgvN9aGio6fOBtoB+0rK8oUbAnegpd1dcXKykpCSdOXNGnTp1Uk5Ojkvz0E9aFkHKR9zepM6ePdvgMVeuXNGKFSskSdOnT1dISEiTc7bmFznHjh0r6dYl7aNHjzZ4zMaNGyXd2j1vzJgxLs/dmmts2bLF+b6usQHehn7S8ryhRsBd6ClNu3r1qkaMGKETJ06oQ4cO2rhx412vyLVmfT7NgE+Ii4szJBmSjEWLFt0xXlVVZYwePdqQZHTr1s0oLi5u/SKbcOHCBSMoKMiQZIwdO/aO8YKCAqNz586GJGPixIkeWeOrr75qcv63337bsNlshiRj+PDhzaoRaAvoJ82zaNEiQ5Jht9vbbI2AJ9BTGldRUWEMHTrUkGT4+fkZb731Vpuqz9cRpHxAZWWlERAQYEgyoqOjjQ4dOhjLly83zp07Z5SUlBi5ublGfHy8IckICAgwPvzwQ0+X3KAXXnjB2WiTk5ONI0eOGBcvXjRycnKM2NhYQ5IRHBxsnDhx4o5zP/74Y+e5b7zxRousER4ebkyaNMn461//ahw7dsy4dOmScfHiRWPPnj3GrFmznCGqc+fOhsPhcOc/DdBq6Ceu95Njx44Ze/fudb5mzpxpSDICAwPr/Xzv3r1GZWWlW2sEvAU9pfGecvPmTWPcuHHOsaysLKOsrKzR17Vr11qkPjSOIOUDDh48aEgybDabkZeXZwQHBzs/TLe/7rnnHiM3N9fT5TaqtrbWmDFjRoO11zWAxup39RcfK2t06dKl0fPqXpGRkcauXbvc8c8BeAT9xPV+MmzYsLv2hLpXQUGBW2sEvAU9pfGeUlBQ4HIPkWT06NGjRepD4/iOlA+o21I0OjpaQ4YM0b/+9S+NHz9e9957r+x2u3r37q1nnnlGJ06c0KhRozxbbBNsNpvWrFmjd955RyNGjFBERITsdrtiY2P11FNPyeFwWK7fyhqvv/665s6dq0ceeUTR0dEKCgpSYGCg7r//fo0YMUIrV67U8ePHNWzYMEs1Ap5EP6FGwJ3oKdTnzWyG0cYf3QzL5s6dq5dfflmjR4/W+++/7+lyAHgx+gkAd6KnwJtxRcoH1O2G873vfc/DlQDwdvQTAO5ET4E3I0j5gLomFRcX5+FKAHg7+gkAd6KnwJsRpNq5M2fO6JtvvpFEkwJgDf0EgDvRU+DtCFLtXN1fevz9/fWd73zHw9UA8Gb0EwDuRE+BtyNItXN1TapXr17q2LGjh6sB4M3oJwDciZ4Cb8eufQAAAABgElekAAAAAMAkghQAAAAAmESQAgAAAACTCFIAAAAAYBJBCgAAAABMIkgBAAAAgEkEKQAAAAAw6X8ArdBjsrO4Ig4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.colors as mcolors\n",
    "import seaborn as sns\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "alphas = [\"0\", \"1e-4\",\"1e-3\",\"1e-2\",\"1e-1\", \"1\"]\n",
    "alphas_n = [0, 1e-4, 1e-3, 1e-2, 1e-1,  1]\n",
    "models = [ 'MNODE05','MNODE10','MNODE20']\n",
    "labels = [ r\"$\\rho=0.05$\",r\"$\\rho=0.10$\", r\"$\\rho=0.20$\"]\n",
    "\n",
    "matplotlib.rcParams['font.family'] = 'Times New Roman'\n",
    "data_pred={}\n",
    "data_rank={}\n",
    "for mod in models:\n",
    "\n",
    "\n",
    "    pred=[]\n",
    "    causal=[]\n",
    "    for a in alphas_n:\n",
    "        if a>1:\n",
    "            print(a)\n",
    "            rmse=np.concatenate([np.load(f\"{mod}_a{a}_pred.npy\"),np.load(f\"{mod}_comp_a{a}_pred.npy\")])\n",
    "            er=np.concatenate([np.load(f\"{mod}_a{a}_causal.npy\"),np.load(f\"{mod}_comp_a{a}_causal.npy\")])\n",
    "        else:\n",
    "            rmse=np.concatenate([np.load(f\"{mod}_a{a}_pred.npy\")[:6],np.load(f\"{mod}_rep1_a{a}_pred.npy\"),np.load(f\"{mod}_rep2_a{a}_pred.npy\")])\n",
    "            er=np.concatenate([np.load(f\"{mod}_a{a}_causal.npy\"),np.load(f\"{mod}_rep1_a{a}_causal.npy\"),np.load(f\"{mod}_rep2_a{a}_causal.npy\")])\n",
    "        pred.append(f\"{np.mean(rmse)}/{np.std(rmse)}\")    \n",
    "        causal.append(f\"{np.quantile(er,0.1)}/{np.quantile(er,0.5)}/{np.quantile(er,0.9)}\")  \n",
    "    data_pred[mod]=pred\n",
    "    data_rank[mod]=causal\n",
    "    \n",
    "\n",
    "def get_values_errors(data_dict):\n",
    "    values = [[float(data_dict[model][alpha].split('/')[0]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors = [[float(data_dict[model][alpha].split('/')[1]) for alpha in range(len(alphas))] for model in models]\n",
    "    return np.array(values), np.array(errors)\n",
    "\n",
    "\n",
    "def get_rank_errors(data_dict): \n",
    "    values = [[float(data_dict[model][alpha].split('/')[1]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors1 = [[float(data_dict[model][alpha].split('/')[0]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors2 = [[float(data_dict[model][alpha].split('/')[2]) for alpha in range(len(alphas))] for model in models]\n",
    "    return np.array(values), np.array(errors1), np.array(errors2)\n",
    "\n",
    "\n",
    "# Custom colors for the bars\n",
    "colors = ['#b3cde3','#b3cde3',  # Very light blue for UVA\n",
    "          '#8c96c6','#8c96c6', '#8c96c6',  # Lighter blue for LP and LPSC\n",
    "          '#8856a7',  # Medium blue for MNODE\n",
    "          '#810f7c', '#810f7c', '#810f7c', '#810f7c', '#810f7c']  # Darker blue for the rest\n",
    "\n",
    "colors_blue = ['#deebf7',   # Very light blue for UVA\n",
    "               '#c6dbef', '#c6dbef',  # Second lightest blue for LP and LPSC\n",
    "               '#9ecae1',  # Third lightest blue for MNODE\n",
    "               '#6baed6', '#6baed6', '#6baed6', '#6baed6', '#6baed6']  # Fourth lightest blue for the rest\n",
    "\n",
    "# Custom colors for ranking bars in light reds\n",
    "colors_red = ['#fee5d9',   # Very light red for UVA\n",
    "              '#fcbba1', '#fcbba1', # Second lightest red for LP and LPSC\n",
    "              '#fc9272',  # Third lightest red for MNODE\n",
    "              '#fb6a4a', '#fb6a4a', '#fb6a4a', '#fb6a4a', '#fb6a4a'] \n",
    "\n",
    "\n",
    "\n",
    "\n",
    "pred_values, pred_errors = get_values_errors(data_pred)\n",
    "rank_values, rank_errors1, rank_errors2 = get_rank_errors(data_rank)\n",
    "\n",
    "# Creating two subplots, one for data_pred and one for data_rank\n",
    "fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(12, 10),facecolor='white')\n",
    "\n",
    "bar_width = 0.1\n",
    "index = np.arange(len(models))\n",
    "\n",
    "def get_gradation_colors(base_color, n):\n",
    "    \"\"\"Generates a gradation of colors from a base color to a lighter shade.\"\"\"\n",
    "    color = mcolors.to_rgb(base_color)\n",
    "    # Adjust the factor to control the rate of color change\n",
    "    factor = 0.9  # Adjust this value as needed to control the darkness\n",
    "    return [mcolors.to_hex((color[0] + (1 - color[0]) * i / (n - 1) * factor, \n",
    "                            color[1] + (1 - color[1]) * i / (n - 1) * factor, \n",
    "                            color[2] + (1 - color[2]) * i / (n - 1) * factor)) for i in range(n)]\n",
    "\n",
    "# Generate gradation colors for each model for predictions\n",
    "gradation_colors_blue = [get_gradation_colors(color, len(alphas)) for color in colors_blue]\n",
    "\n",
    "# Generate gradation colors for each model for rankings\n",
    "gradation_colors_red = [get_gradation_colors(color, len(alphas)) for color in colors_red]\n",
    "\n",
    "# Plotting bars with gradation colors for data_pred\n",
    "for i, model in enumerate(models):\n",
    "    for j, alpha in enumerate(alphas):\n",
    "        ax1.bar(index[i] + j*bar_width, pred_values[i][j], yerr=pred_errors[i][j], \n",
    "                color=gradation_colors_blue[i][j], edgecolor='black', ecolor='#3b3b3b', \n",
    "                width=bar_width, label=f'α={alpha}' if i == 0 else \"\")\n",
    "\n",
    "# Plotting bars with gradation colors for data_rank\n",
    "for i, model in enumerate(models):\n",
    "    for j, alpha in enumerate(alphas):\n",
    "        ax2.bar(index[i] + j*bar_width, rank_values[i][j], yerr=[[rank_values[i][j]-rank_errors1[i][j]],\\\n",
    "                                                                 [rank_errors2[i][j]-rank_values[i][j]]], \n",
    "                color=gradation_colors_red[i][j], edgecolor='black', ecolor='#3b3b3b', \n",
    "                width=bar_width, label=f'α={alpha}' if i == 0 else \"\")\n",
    "\n",
    "\n",
    "# Customizing the plot for aesthetics\n",
    "ax1.set_title('Prediction RMSE (Hybrid Loss)', fontsize=26)\n",
    "ax1.set_ylim([0,100])\n",
    "ax2.set_ylim([0,1])\n",
    "ax2.set_title('Causal Error Rate 10/50/90-th Percentile (Hybrid Loss)', fontsize=26)\n",
    "ax1.set_xticks(index + bar_width * 2.5)\n",
    "ax2.set_xticks(index + bar_width * 2.5)\n",
    "ax1.set_xticklabels(labels, rotation=0, fontsize=26, color='black')\n",
    "ax2.set_xticklabels(labels, rotation=0, fontsize=26, color='black')\n",
    "ax1.tick_params(axis='y', labelsize=26)  # Adjust the labelsize as needed for the first subplot\n",
    "ax2.tick_params(axis='y', labelsize=26)  # Adjust the labelsize as needed for the second subplot\n",
    "\n",
    "# Add a dashed light gray horizontal line at y=0.33 in the ranking subplot\n",
    "ax2.axhline(y=0.66, color='gray', linestyle='--', linewidth=1, zorder=0)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"final_plots/corruption_complete2.pdf\", format='pdf', dpi=800)  # You can adjust the filename, format, and dpi as needed\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "4f848b67-28c2-4b5c-842b-454b350a8e44",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.colors as mcolors\n",
    "import seaborn as sns\n",
    "import matplotlib\n",
    "import numpy as np\n",
    "\n",
    "alphas = [\"0\", \"1e-4\", \"1e-3\", \"1e-2\", \"1e-1\", \"1\"]\n",
    "alphas_n = [0, 1e-4, 1e-3, 1e-2, 1e-1, 1]\n",
    "models = [\"UVA\",\"LP\", \"LPSC\", \"MNODE\",\"BNODE\", \"TCN\",\"Tran\",\"LSTM\",\"S4D\"]\n",
    "labels = models\n",
    "\n",
    "\n",
    "# Custom colors for the bars\n",
    "colors = ['#b3cde3',  # Very light blue for UVA\n",
    "          '#8c96c6', '#8c96c6',  # Lighter blue for LP and LPSC\n",
    "          '#8856a7',  # Medium blue for MNODE\n",
    "          '#810f7c', '#810f7c', '#810f7c', '#810f7c', '#810f7c']  # Darker blue for the rest\n",
    "\n",
    "data_pred={}\n",
    "data_rank={}\n",
    "plot_comp=False\n",
    "for mod in models:\n",
    "    mod2=mod+\"OOD\"\n",
    "    \n",
    "    pred=[]\n",
    "    causal=[]\n",
    "    for a in alphas_n:\n",
    "        #if a==1 and mod==\"TCN\":\n",
    "        #    mod2=\"TCNOOD_rep0\"\n",
    "        if a<1:\n",
    "            if mod==\"LP\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_reduced_a{a}_pred.npy\")[:6],np.load(f\"{mod2}_rep1_a{a}_pred.npy\"),np.load(f\"{mod2}_rep2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_reduced_a{a}_causal.npy\"),np.load(f\"{mod2}_rep1_a{a}_causal.npy\"),np.load(f\"{mod2}_rep2_a{a}_causal.npy\")])\n",
    "            elif mod==\"UVA\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_full_a{a}_pred.npy\")[:6],np.load(f\"{mod2}_repboth_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_full_a{a}_causal.npy\"),np.load(f\"{mod2}_repboth_a{a}_causal.npy\")])\n",
    "            elif mod==\"LPSC\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_reduced_a{a}_pred.npy\")[:6],np.load(f\"{mod2}_rep1_a{a}_pred.npy\"),np.load(f\"{mod2}_rep2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_reduced_a{a}_causal.npy\"),np.load(f\"{mod2}_rep1_a{a}_causal.npy\"),np.load(f\"{mod2}_rep2_a{a}_causal.npy\")])\n",
    "            elif mod=='BNODE':\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"{mod2}_rep1_a{a}_pred.npy\"),np.load(f\"{mod2}_rep2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\"),np.load(f\"{mod2}_rep1D_a{a}_causal.npy\"),np.load(f\"{mod2}_rep2D_a{a}_causal.npy\")])\n",
    "            else:\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"{mod2}_rep1_a{a}_pred.npy\"),np.load(f\"{mod2}_rep2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\"),np.load(f\"{mod2}_rep1_a{a}_causal.npy\"),np.load(f\"{mod2}_rep2_a{a}_causal.npy\")])\n",
    "        else:\n",
    "            if mod==\"LPSC\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_reduced_a{a}_pred.npy\")[:6],np.load(f\"LPSCOOD_comp1_a{a}_pred.npy\"),np.load(f\"LPSCOOD_comp2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_reduced_a{a}_causal.npy\")[:6],np.load(f\"LPSCOOD_comp1_a{a}_causal.npy\"),np.load(f\"LPSCOOD_comp2_a{a}_causal.npy\")])\n",
    "            if mod==\"LP\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_reduced_a{a}_pred.npy\")[:6],np.load(f\"LPOOD_comp1_a{a}_pred.npy\"),np.load(f\"LPOOD_comp2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_reduced_a{a}_causal.npy\")[:6],np.load(f\"LPOOD_comp1_a{a}_causal.npy\"),np.load(f\"LPOOD_comp2_a{a}_causal.npy\")])\n",
    "            if mod==\"Tran\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"TranOOD_comp_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"TranOOD_comp_a{a}_causal.npy\")])\n",
    "            if mod==\"BNODE\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"BNODEOOD_comp1_a{a}_pred.npy\"),np.load(f\"BNODEOOD_comp2_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"BNODEOOD_comp1D_a{a}_causal.npy\"),np.load(f\"BNODEOOD_comp2D_a{a}_causal.npy\")])\n",
    "            if mod==\"MNODE\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"MNODEOOD_comp_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"MNODEOOD_comp_a{a}_causal.npy\")])\n",
    "            if mod==\"UVA\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_full_a{a}_pred.npy\")[:6],np.load(f\"{mod2}_repboth_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_full_a{a}_causal.npy\"),np.load(f\"{mod2}_repboth_a{a}_causal.npy\")])\n",
    "            if mod==\"LSTM\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"LSTMOOD_comp_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"LSTMOOD_comp_a{a}_causal.npy\")])\n",
    "            if mod==\"TCN\":\n",
    "                rmse=np.load(f\"TCNOOD_comp_a{a}_pred.npy\")\n",
    "                er=np.load(f\"TCNOOD_comp_a{a}_causal.npy\")\n",
    "                #print(er)\n",
    "            if mod==\"S4D\":\n",
    "                rmse=np.concatenate([np.load(f\"{mod2}_a{a}_pred.npy\")[:6],np.load(f\"S4DOOD_comp_a{a}_pred.npy\")])\n",
    "                er=np.concatenate([np.load(f\"{mod2}_a{a}_causal.npy\")[:6],np.load(f\"S4DOOD_comp_a{a}_causal.npy\")])\n",
    "        pred.append(f\"{np.mean(rmse)}/{np.std(rmse)}\")    \n",
    "        causal.append(f\"{np.quantile(er,0.1)}/{np.quantile(er,0.5)}/{np.quantile(er,0.9)}\")\n",
    "    data_pred[mod]=pred\n",
    "    data_rank[mod]=causal\n",
    "\n",
    "# Custom colors for the bars\n",
    "colors = ['#b3cde3', # Very light blue for UVA\n",
    "          '#8c96c6','#8c96c6',   # Lighter blue for LP and LPSC\n",
    "          '#8856a7',  # Medium blue for MNODE\n",
    "          '#810f7c', '#810f7c', '#810f7c', '#810f7c', '#810f7c']  # Darker blue for the rest\n",
    "\n",
    "colors_blue = ['#deebf7',   # Very light blue for UVA\n",
    "               '#c6dbef', '#c6dbef',  # Second lightest blue for LP and LPSC\n",
    "               '#9ecae1',   # Third lightest blue for MNODE\n",
    "               '#6baed6', '#6baed6', '#6baed6', '#6baed6', '#6baed6']  # Fourth lightest blue for the rest\n",
    "\n",
    "# Custom colors for ranking bars in light reds\n",
    "colors_red = ['#fee5d9',   # Very light red for UVA\n",
    "              '#fcbba1', '#fcbba1',  # Second lightest red for LP and LPSC\n",
    "              '#fc9272',  # Third lightest red for MNODE\n",
    "              '#fb6a4a', '#fb6a4a', '#fb6a4a', '#fb6a4a', '#fb6a4a'] \n",
    "\n",
    "def get_values_errors(data_dict):\n",
    "    values = [[float(data_dict[model][alpha].split('/')[0]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors = [[float(data_dict[model][alpha].split('/')[1]) for alpha in range(len(alphas))] for model in models]\n",
    "    return np.array(values), np.array(errors)\n",
    "\n",
    "\n",
    "def get_rank_errors(data_dict): \n",
    "    values = [[float(data_dict[model][alpha].split('/')[1]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors1 = [[float(data_dict[model][alpha].split('/')[0]) for alpha in range(len(alphas))] for model in models]\n",
    "    errors2 = [[float(data_dict[model][alpha].split('/')[2]) for alpha in range(len(alphas))] for model in models]\n",
    "    return np.array(values), np.array(errors1), np.array(errors2)\n",
    "\n",
    "\n",
    "pred_values, pred_errors = get_values_errors(data_pred)\n",
    "rank_values, rank_errors1, rank_errors2 = get_rank_errors(data_rank)\n",
    "\n",
    "# Creating two subplots, one for data_pred and one for data_rank\n",
    "fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(12, 10),facecolor='white')\n",
    "\n",
    "bar_width = 0.1\n",
    "index = np.arange(len(models))\n",
    "\n",
    "def get_gradation_colors(base_color, n):\n",
    "    \"\"\"Generates a gradation of colors from a base color to a lighter shade.\"\"\"\n",
    "    color = mcolors.to_rgb(base_color)\n",
    "    # Adjust the factor to control the rate of color change\n",
    "    factor = 0.9  # Adjust this value as needed to control the darkness\n",
    "    return [mcolors.to_hex((color[0] + (1 - color[0]) * i / (n - 1) * factor, \n",
    "                            color[1] + (1 - color[1]) * i / (n - 1) * factor, \n",
    "                            color[2] + (1 - color[2]) * i / (n - 1) * factor)) for i in range(n)]\n",
    "\n",
    "# Generate gradation colors for each model for predictions\n",
    "gradation_colors_blue = [get_gradation_colors(color, len(alphas)) for color in colors_blue]\n",
    "\n",
    "# Generate gradation colors for each model for rankings\n",
    "gradation_colors_red = [get_gradation_colors(color, len(alphas)) for color in colors_red]\n",
    "\n",
    "# Plotting bars with gradation colors for data_pred\n",
    "for i, model in enumerate(models):\n",
    "    for j, alpha in enumerate(alphas):\n",
    "        ax1.bar(index[i] + j*bar_width, pred_values[i][j], yerr=pred_errors[i][j], \n",
    "                color=gradation_colors_blue[i][j], edgecolor='black', ecolor='#3b3b3b', \n",
    "                width=bar_width, label=f'α={alpha}' if i == 0 else \"\")\n",
    "\n",
    "# Plotting bars with gradation colors for data_rank\n",
    "for i, model in enumerate(models):\n",
    "    for j, alpha in enumerate(alphas):\n",
    "        ax2.bar(index[i] + j*bar_width, rank_values[i][j], yerr=[[rank_values[i][j]-rank_errors1[i][j]],\\\n",
    "                                                                 [rank_errors2[i][j]-rank_values[i][j]]], \n",
    "                color=gradation_colors_red[i][j], edgecolor='black', ecolor='#3b3b3b', \n",
    "                width=bar_width, label=f'α={alpha}' if i == 0 else \"\")\n",
    "\n",
    "\n",
    "# Customizing the plot for aesthetics\n",
    "ax1.set_title('Prediction RMSE (Hybrid Loss)', fontsize=26)\n",
    "ax1.set_ylim([0,100])\n",
    "ax2.set_title('Causal Error Rate 10/50/90-th Percentile (Hybrid Loss)', fontsize=26)\n",
    "ax1.set_xticks(index + bar_width * 2.5)\n",
    "ax2.set_xticks(index + bar_width * 2.5)\n",
    "ax1.set_xticklabels(models, rotation=45, fontsize=26, color='black')\n",
    "ax2.set_xticklabels(models, rotation=45, fontsize=26, color='black')\n",
    "ax1.tick_params(axis='y', labelsize=26)  # Adjust the labelsize as needed for the first subplot\n",
    "ax2.tick_params(axis='y', labelsize=26)  # Adjust the labelsize as needed for the second subplot\n",
    "\n",
    "# Add a dashed light gray horizontal line at y=0.33 in the ranking subplot\n",
    "ax2.axhline(y=0.66, color='gray', linestyle='--', linewidth=1, zorder=0)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"final_plots/OOD_complete2.pdf\", format='pdf', dpi=800)  # You can adjust the filename, format, and dpi as needed\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5caf6ab9-f5fa-41c6-84f9-731f21171127",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}