{ "cells": [ { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "# A deep neural network of energy absorbing structures" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "In this notebook you will implement a deep neural network to represent the response of foam-filled aluminium extrusions.\\\n", "\n", "In our dataset we have three inputs which are the width of the squared aluminium profile, the yield strength of the aluminium alloy as well as the density of the foam. As outputs we have the maximum force and the mean force during crushing of the profile.\n", "\n", "You have to go through the following steps:\n", "1. Import the dataset Data_Example_3.txt, the dataset contains 3 inputs and 2 outputs\n", "2. Scale the data using sklearn\n", "3. Split the data into training and validation using sklearn (use a 80/20 split)\n", "4. Build an ANN with two hidden layers consisting of 10 neurons with the ReLu activation function\n", "5. Train the ANN for 100 epochs using the adam optimizer\n", "6. Plot the training and validation losses, can you comment on them?\n", "7. Load the dataset called Data_Example_3_interpolation.txt and Data_Example_3_Extrapolation.txt\n", "8. Evaluate the trained ANN on these two additional datasets, can you comment on the results?\n", "9. What can you do to improve the results?\n" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "# We import the necessary packages\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import warnings\n", "warnings.filterwarnings('ignore')\n", "import tensorflow as tf\n", "#----------------------------------------------------------------------------------------------------------------\n", "# Import SciKitLearn for normalization and splitting of data\n", "#----------------------------------------------------------------------------------------------------------------\n", "from sklearn.model_selection import train_test_split\n", "from sklearn import preprocessing" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "We import the data from the file 'Data_Example_5.txt'. Since this file uses commas to delimit columns we need to specify it to the numpy loadtxt function." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "data = np.loadtxt('Data_Example_3.txt',delimiter=',')" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "Next we apply a min/max scaler from Scikit to the dataset." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "scaler = preprocessing.MinMaxScaler().fit(data)\n", "scaled_data = scaler.transform(data)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "We then split the data between input and output using the Scikit function train_test_split." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "X_scaled = scaled_data[:,0:3]\n", "Y_scaled = scaled_data[:,3:]\n", "#\n", "X_train, X_val, Y_train, Y_val = train_test_split(X_scaled, Y_scaled,test_size = 0.2,random_state = 1)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "We can now build the ANN model." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "model = tf.keras.models.Sequential()\n", "model.add(tf.keras.layers.Dense(units=10, activation='relu',input_shape=(3,)))\n", "model.add(tf.keras.layers.Dense(units=10, activation='relu'))\n", "model.add(tf.keras.layers.Dense(units=2, activation='linear'))\n", "model.compile(loss='mse', optimizer='adam')" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "We train the ANN for 100 epochs using the Adam optimizer." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "history = model.fit(X_train, Y_train, epochs=100,validation_data=(X_val,Y_val), verbose=0)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "We plot the results from the training in terms of training and validation losses." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAr4AAAHACAYAAABER2tZAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjEsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvc2/+5QAAAAlwSFlzAAAPYQAAD2EBqD+naQAAds5JREFUeJzt3Qd4FNXXBvB3N733QgihJZRQQ+i9d6SIIEWKiCIBUT4L6h9FsYvYQLHQpAtKk95D7z20QEgCgVTSe7Lfc++QQCCBAEm25P09z7izs7OzdxmBk8u556g0Go0GREREREQGTq3tARARERERlQUGvkRERERULjDwJSIiIqJygYEvEREREZULDHyJiIiIqFxg4EtERERE5QIDXyIiIiIqFxj4EhEREVG5YKztAei63NxcREREwMbGBiqVStvDISIiIqIHiH5sSUlJ8PDwgFpd9LwuA9/HEEFvpUqVtD0MIiIiInqM8PBweHp6Fvk6A9/HEDO9eb+Qtra2pf55WVlZ2Lp1K7p27QoTE5NS/zwqHbyPhoH30TDwPuo/3kPDkFWK9zExMVFOVObFbUVh4PsYeekNIugtq8DX0tJSfhZ/c+sv3kfDwPtoGHgf9R/voWHIKoP7+Li0VC5uIyIiIqJygYEvEREREZULDHyJiIiIqFxgji8REREZZHmr7Oxs5OTkaHsodF+Or7GxMdLT05/4vhgZGcn3PmtpWQa+REREZFAyMzNx69YtpKamanso9MAPI+7u7rJS1tMEsGJhXIUKFWBqaoqnxcC3CLNnz5Ybf1IkIiLSr8ZTISEhcoZQNDMQQRIbUOnOvUlOToa1tfUjm0wUFjCLH2aio6PlvfXx8Xmi99+PgW8RAgIC5CbqwtnZ2Wl7OERERFQMIkASAZao6SpmCEl35Obmyvtjbm7+xIGrhYWFLIEWGhqaf42nwcVtREREZHCedkaQDPue8v8KIiIiIioXGPgSERERGZgqVarghx9+KPb5u3fvlrnQ8fHxMGTM8SUiIiLSAe3bt0fDhg2fKGAtytGjR2FlZVXs81u2bCkrYRj6uiYGvkRERER6QFQ3ENWmRD3bx3FxcXmia5uamspSY4aOqQ5EREREWjZq1Cjs2bMHP/74o0w5ENuCBQvk46ZNm+Dv7w8zMzPs27cPV69eRd++feHm5iZLgzVp0gTbt29/ZKqDuM6ff/6J/v37y2oXoiTYunXrikx1EJ9tb2+PLVu2oHbt2vJzunfvLmeF84gGIW+88YY8z8nJCe+99x5GjhyJfv36QVcx8NUhtxLS8PycQ/jqtJG2h0JERGRQM6WpmdllvonPLS4R8LZo0QJjx46VwaXYREk2YcqUKfjqq69w4cIF1K9fX9bC7dmzJ3bs2IGTJ0/KgLRPnz4ICwt75Gd88sknGDRoEM6cOSPfP2zYMMTFxRV5vmgAMmPGDCxatAiBgYHy+m+//Xb+619//TWWLFmC+fPnY//+/bIE7Jo1a6DLmOqgQyxMjHDmZqL4uQwZ2bkwMdH2iIiIiPRfWlYOfD/aUuafG/RpN1iaFi/UErm1It1AzMbmpRxcvHhRPn766afo0qVL/rmOjo5o0KBB/vPp06dj9erVcgZ3woQJj5xVHjJkiNz/4osv8NNPP+HIkSMycC6qxfCcOXNQvXp1+VxcW4wlz88//4z3339fziILs2bNwsaNG6HLOOOrQ+wsTGBipHSXiUvJ1PZwiIiISAc0bty4wHMx4ytmXkUKgkgzEGkIYjb4cTO+9evXz98XC99sbW0RFRVV5PkiCM8LegXRLjjv/ISEBERGRqJp06b5r4tueSIlQ5dxxleHiNwaRytTRCZmICY5A17ONtoeEhERkUH8i6qYfdXG55aEB6sziKB327ZtMg3B29tbdjUbOHCg7Gj2KCYP/FOyiDtEN7UnOf9J0jd0EQNfHeNsnRf4csaXiIioJIiArbgpB9okUh1E1YbHEfm0Im0hL8VAzABfv34dZcnOzk4urhNl09q2bSuPibGfOHFClmTTVbr/f0E542xlBiCJgS8REVE5IyoxHD58WAaxIn2hqNlYUZHh33//lQvaRFA/derUR87clpaJEyfiyy+/lLPOtWrVkjm/d+7ckWPSVczx1TFO1qbyMTY5Q9tDISIiojIkUhhEnqyvr6+sw1tUzu7MmTPh4OAgm06I4Ldbt25o1KhRmY/3vffek4vlRowYIStSiGBdjMXc3By6ijO+OpjqIMRwcRsREVG5UqNGDRw8eLDAMZHSUNjM8M6dOwscCwgIKPD8wdSHwnJz4+9rTyy6xt1/jvjcBz9b1Oe9/xzRSEPM8opNELPOYsGdKJmmqxj46hhna5HqAKY6EBERkU4LDQ3F1q1b0a5dO2RkZMhyZiEhIRg6dCh0FVMddIyTFVMdiIiISPep1WrZ4U10jmvVqhXOnj0rO8iJWV9dxRnfIsyePVtuxVldWRo5vpzxJSIiIl1WqVIlWWFCn3DGtwgiVyYoKEiW6ShLznkzvszxJSIiIipRDHx1dHHbndQsZOWUfWkSIiIiIkPFwFfH2FuaQgVlxSTbFhMRERGVHAa+OsZIrYL13Q6Bom0xEREREZUMBr46yCY/8OWMLxEREVFJYeCrg2xMlFSHmCTO+BIRERGVFAa+Oj3jy8CXiIiIikd0dPvhhx/yn6tUKqxZs6bI869fvy7POXXq1DN9bkldpyywjq8OYuBLREREz+rWrVtwcHAo0WuOGjVKtjq+P6AW9XzFZzk7O0PXMfDV5VQH5vgSERHRU3J3dy+TzzEyMiqzz3pWTHXQQTZKKV/O+BIREZUTv//+Ozw8PJCbW7CGf9++ffHyyy/j6tWrct/NzQ3W1tayTbBoD/woD6Y6HDlyBH5+fjA3N0fjxo1x8uTJAueLbrVjxoxB1apVYWFhgZo1a+LHH3/Mf33atGlYuHAh1q5dK68ttt27dxea6rBnzx40bdoUZmZmqFChAqZMmYLs7Oz819u3b4833ngD7777LhwdHWXgLK5f2jjjq8OpDtFc3EZERPTsNBogK7XsP9fEUkSfxTr1hRdewMSJE7Fr1y506tRJHouLi8PmzZuxceNGJCcno2fPnvj8889lMPnXX3+hT58+uHTpEry8vB57ffH+3r17o0uXLli8eDFCQkIwadKkAueIoNvT0xMrV66Ek5MTDhw4gFdffVUGroMGDcLbb7+NCxcuIDExEfPnz5fvEUFrREREgevcvHlTjlWkRYhxXrx4EWPHjpXjfuutt/LPE0H05MmTcfjwYRw8eFCe36pVKznG0sLAVwcx1YGIiKgEiaD3C4+y/9wPIgBTq2KdKnJxe/TogaVLl+YHvqtWrZJ5sx06dIBarUaDBg3yz58+fTpWr16NdevWYcKECY+9vrhubm4u5s6dK2d869Spgxs3buD111/PP8fExASffPJJ/nMx8ysC0r///lsGvmKmWcwEZ2RkPDK14ZdffpF5v7NmzZIzwbVq1ZLB8XvvvVcg2K5fvz4+/vhjue/j4yPP37FjR6kGvkx10OEZ37iUDOTkKkEwERERGbZhw4bhn3/+kYGlsGTJErz44osy6BUztmLGtXbt2rC3t5dBqJh9DQsLK9a1xbn169eXQW+eFi1aPHTe7Nmz4e/vDxcXF/kZIgWjuJ9x/2eJa4ugN4+YyRXfQcwG5xHjuZ+YWY6KikJp4oyvDsrr3CZi3jupmXC2NtP2kIiIiPSXSDkQs6/a+NwnIFIXNBoNNmzYIHN49+7di++//16+JoLebdu2YcaMGfD29pYzrwMHDkRmZsn96/Dy5cvl53z33XcycLWxscG3334rUxFKg5hhvp8IlB/McS5pDHx1kJEKcLA0wZ3ULLnAjYEvERHRMxAzj8VMOdAmMRs7YMAAOdMbHBwsF5c1atRIvrZ//36ZA9u/f3/5XMyeikVlxSVmihctWoT09PT8Wd9Dhw4VOEd8RsuWLTF+/Pj8Y2JR3f1MTU3lIrjHfZaYuRZBfN6sr7i2CKQrVqwIbWKqg45yslJKO8QkMc+XiIioPKU7iBnfefPmyf08Igf233//lZUTTp8+jaFDhz7R7Kg4X6VSyUVmQUFBcsGcmD2+n/iMY8eOYcuWLbh8+TKmTp2Ko0ePPtQk48yZM3JRXUxMDLKysh76LBE4h4eHy8V6YmGbqAIhcnnFwjaRtqFNDHx1lLP13cCXJc2IiIjKjY4dO8pKCSKwFMFqnpkzZ8oFcGJGVqREdOvWLX82uDhEvu769etx9uxZWdLsww8/xNdff13gnNdee03OOA8ePBjNmjVDbGxsgdlfQQTOYiZalEMTecBiJvdBYlZXBNaifJpYkDdu3DhZJk18prYx1UFHOd1Nb2DgS0REVH6IGdEHy4PlzbTu3LmzwLGAgIACzx9MfRCpBvdr3rz5Q22F7z9HlBsTZcrySpXl+fLLL/P3RbC7devWh8b34Ge1a9dOBr73u3+GWtT/fdCj2iuXFM746viMbzQDXyIiIqISwcBXRznfzfGNZS1fIiIiohLBwFdHMdWBiIiIqGQx8NVRXNxGREREVLIY+Op64MtyZkREREQlgoFvEUTLPl9fX9k5RRvymlbEpmQ8tFKSiIiIHo1/dxoeTQncUwa+RRAlQkSB5wcLN5cVx7uL27JyNEhIe7g4NBERERXdBjc1NVXbQ6ESlndPH2x1/CRYx1dHmRmrYWtujMT0bJnna2+pBMJERERUNCMjI9jb2yMqKko+t7S0zG+bS9ol6vhmZmbKtslP0sFNzPSKoFfcU3FvxT1+Wgx8dZizjZkMfKOTMuHtqu3REBER6Qd3d3f5mBf8km7QaDRIS0uDhYXFU/0wIoLevHv7tBj46jCR53stOoWVHYiIiJ6ACKoqVKgAV1dXZGUxXVBXZGVlITAwEG3btn3idAVx/rPM9OZh4KvDWNKMiIjo6YlAqSSCJSoZ4l5kZ2fD3Nz8mfJ0nwUXt+mwvMoODHyJiIiInh0DX30IfFnLl4iIiOiZMfDVYZzxJSIiIio5DHx1GHN8iYiIiEoOA18dL2cmxCQz1YGIiIjoWTHw1WEud1MdopPZtpiIiIjoWTHw1YMc38zsXCRnZGt7OERERER6jYGvDrMwNYKVqVJ/kOkORERERM+Gga/e5PlygRsRERHRs2Dgqze1fBn4EhERET0LBr46jiXNiIiIiEoGA189mfGNZo4vERER0TNh4Kvj2L2NiIiIqGQw8NWXxW3M8SUiIiJ6Jgx8dZyzFXN8iYiIiEoCA18dx7bFRERERCWDga+OY44vERERUclg4KtrIs/B486Rh8qZpWbmIDWTbYuJiIiInhYDX10SfgQmf7ZHw/B5QHa6PGRtZgwzY+U2xSQx3YGIiIjoaTHw1SUVG0Nj6wmTnFSoLm+Wh1Qq1X21fJnuQERERPS0GPjqErUaufUGKbtnlheywI2BLxEREdHTYuCrY3LrvSAfVdd2AUmRct+FbYuJiIiInhkDX13j5IM4y+pQaXKAsyvlobxUh1iWNCMiIiJ6agx8dVC4Y2tl57SS7sCSZkRERETPjoGvDrrp0AwaI1Mg8ixw+2x+SbMbd9K0PTQiIiIivcXAVwdlGVtD49NdeXJqGZpUdZS7+67E4E4K0x2IiIiIngYDXx2VV90BZ/9GHTcr1PGwRWZOLlafvKntoRERERHpJQa+OkpTvRNg6QykRANXd2Bwk0ry+N/HwqHRaLQ9PCIiIiK9w8BXVxmZAHdLm+H0MvRtUBGmxmpcvJ2EMzcStD06IiIiIr3DwFeXNRyiPF7cCDtVMnrUdZdPVxwL1+64iIiIiPQQA19d5l4fcPUFcjKA82swuLGS7rD+VATSMnO0PToiIiIivWLwgW94eDjat28PX19f1K9fHytXKk0h9IJKBTS4O+t7ehmaV3NCJUcLJGVkY9O5W9oeHREREZFeMfjA19jYGD/88AOCgoKwdetWvPnmm0hJSYHeqD8IUKmB8MNQ37mGQf7KrO+Ko0x3ICIiInoSBh/4VqhQAQ0bNpT77u7ucHZ2RlxcHPSGjTtQvaOyf3oZBjb2lBPBh0PicD1GjwJ4IiIiovIe+AYGBqJPnz7w8PCASqXCmjVrHjpn9uzZqFKlCszNzdGsWTMcOXLkqT7r+PHjyMnJQaVKyqyp3mg4THk8vgAVLIG2Pi75pc2IiIiIqHiMoWUi7aBBgwZ4+eWXMWDAgIdeX7FiBSZPnow5c+bIoFekLXTr1g2XLl2Cq6urPEfM6GZnZz/0XpHaIAJqQczyjhgxAn/88ccjx5ORkSG3PImJifIxKytLbqUt7zMKfJZ3dxjbVYIqIRw5Jxbjeb/u2HM5GquO38DE9lVhbKT1n1+oOPeR9A7vo2HgfdR/vIeGIasU72Nxr6nS6FA3BDHju3r1avTr1y//mAh2mzRpglmzZsnnubm5csZ24sSJmDJlSrGuKwLZLl26YOzYsXjppZceee60adPwySefPHR86dKlsLS0hLZUi9qKejcXI9nMDVtqfo2PTpggJVuFsbVyUNdBZ24hERERUZlLTU3F0KFDkZCQAFtbW92d8X2UzMxMmZ7w/vvv5x9Tq9Xo3LkzDh48WKxriLh+1KhR6Nix42ODXkF8lphhvn/GVwTaXbt2feQvZEn+xLJt2zYZqJuYmNx7IbMdNLM2wDotEn19NDhnXAXzD4TiuqoC3u2p5DCT7ijyPpJe4X00DLyP+o/30DBkleJ9zPsX+sfR6cA3JiZG5uS6ubkVOC6eX7x4sVjX2L9/v0yXEKXM8vKHFy1ahHr16hV6vpmZmdweJG5QWf5me+jzTOyBJmOBwG9gfPBnDHlurQx8d12KRnx6LlxsHh4zaV9Z/39DpYP30TDwPuo/3kPDUBr3sbjX0+nAtyS0bt1apkcYhKavAgd+AiJOoEb6GTSsZI9T4fGYvSsY056ro+3REREREek0nV4VJUqPGRkZITIyssBx8VyUJit3rF3uVXjY/yP+r2sNubvw4HWcCLuj3bERERER6TidDnxNTU3h7++PHTt25B8Ts7fieYsWLVAutQhQGlpc2Yo2tlEY0KgixPLE9/85i8xsA5nZJiIiIjLEwDc5ORmnTp2SmxASEiL3w8LC5HOx0EyUIFu4cCEuXLiA119/XZZAGz16NMolp+pA7eeU/QM/43+9fOFoZYpLkUn4bc9VbY+OiIiISGdpPfA9duwY/Pz85JYX6Ir9jz76SD4fPHgwZsyYIZ+Ler0iKN68efNDC95Kmmia4evrK0up6ZxWbyiPZ1fCMTsKH/X2lU9/3hmMq9HJ2h0bERERkY7SeuDbvn17WXLswW3BggX550yYMAGhoaGyHu/hw4dlbd/SFhAQgKCgIBw9ehQ6p6I/UKUNkJsNHPoVfRt6oF0NF2Tm5OL9f88iN5d1fYmIiIh0LvClp9TqTeXx+AKo0uPxWb+6sDAxwpGQOKxgK2MiIiKihzDw1VfenQDXOkBmspz1reRomV/l4YuNFxCVmK7tERIRERHpFAa++kqlAtq9q+wfnA2kxGB0q6qo72mHpPRsTFt/XtsjJCIiItIpDHz1majuUKGBMuu773sYqVX4akB9+bjx7G3suRyt7RESERER6QwGvvpMrQY6KdUvcOQPIOEGfD1sMaplFXnok/XnWduXiIiI6C4GvvpYzux+1TsBlVsBORnAnm/koUmdfeBsbYpr0SlYeOC6tkdIREREpBMY+OpjObMHc33zZn1PLgZigmFrboJ3u9eSh37ccQVRSVzoRkRERMTA1xB4NQd8ugGaHGD3F/LQwEaeaOBph+SMbHyz+ZK2R0hERESkdQx8DUXH/ymP5/4Bbp+FWq3CtOfqyEOrjt/AibA72h0fERERkZYx8DUUFeoDdZ9X9ndMlw9+Xg54wd9T7k9bd54d3YiIiKhcY+BrSNp/AKiMgCtbgLBD8pDI9bUxM8aZGwly5peIiIiovGLga0icvQG/Ycr+jk8BjQYuNmayyoPw9eaLSEjL0u4YiYiIiLSEga+hafceYGQGhO4HQgLloREtqqC6ixViUzLx844r2h4hERERkVYw8NX3Or4PsvME/Ecq+3fr+poaqzG1t6/cX3I4DPGpmdocIREREZFWMPDV9zq+hWn1JmBkCoTuA67vk4fa1XCBbwVbpGXlYOmRMG2PkIiIiKjMMfA1RHYVAb/hyv6er+WDSqXCK22qyn3RzY2tjImIiKi8YeBrqFpPBtQmSp5v6EF5qHd9D7jamCEyMQMbzkZoe4REREREZYqBr6GyrwQ0HKrsB97L9R3Zsorc/3NvCDQa1vUlIiKi8oOBryFrM1mp63t1JxCu5CoPbeoFcxM1zkck4tC1OG2PkIiIiKjMMPA1ZA5VgAZDCsz6OliZYuDdbm5z913T5uiIiIiIyhQDX0PX9v/udnPbCtw8IQ+93EpZ5Lb9QhSuRSdreYBEREREZYOBr6FzrAbUH1Sgrm81F2t0quUq9+fvv67N0RERERGVGQa+5UEbMeurBi5vAm6dlofG3C1ttvJ4OBtaEBERUbnAwNfQOrcVxtkHqPu8sr93pnxoUc1JNrRIz8qV3dyIiIiIDB0DX0Ps3FaYlhOVx8ubgYxkNrQgIiKicoeBb3nhXh9wqApkpwPB2ws0tIhKysDGs7e0PUIiIiKiUsXAt7xQqYDafZT9C+vyG1oMbeYl9/85cUOboyMiIiIqdQx8yxPfvsrj5S1AVrrc7e9XUT7uD45BVJJyjIiIiMgQMfAtTzwaAbYVgcxk4NpueaiykxX8vOyRqwHWn2a6AxERERkuBr7liVoN1OpdIN1B6NdQmfVde+qmtkZGREREVOoY+JY3vs8pj5c2AjlZcrdX/QowUqtw5kYCO7kRERGRwWLgW954tQAsnYG0O8D1ffKQs7UZ2vg4y/01pyK0PEAiIiKi0sHAt7xRGwG1ej0y3UGj0WhrdERERESlhoFveVT7brrDhf+A3By528XXDRYmRgiNTcXJ8Hjtjo+IiIioFDDwLQ8tix9UtS1gZgekRAHhR+QhKzNjdKvjJvfXnuQiNyIiIjI8DHzLS8vi+xmbAjW7K/sX1ucf7nu3pu9/Z24hK4ctjImIiMiwMPBFeU93WA/czelt4+0MJytTxKZkYl9wjHbHR0RERFTCGPiWV96dABNLICEMiDgpDxkbqdG7fgW5z3QHIiIiMjQMfMsrEwvAp0uR6Q5bgyKRmpmtrdERERERlTgGvuVZfrrDuvx0B79K9qjsZInUzBxsC4rU7viIiIiIShAD3/LMpytgZArEBgPRF+UhlUqFvndr+q5hugMREREZEAa+5Zm5LVC9o7J/fnX+4X4NPeRj4JUYRCdlaGt0RERERCWKgW95V2eA8nh2VX66QzUXa/h52SMnV4NlR8K0Oz4iIiKiEsLAt7yr1RMwNgfirgK3TuUfHtmiinxccjiUNX2JiIjIIDDwLe/MbICaPe7N+t7Vs14FuNiYITIxA5vP3dbe+IiIiIhKCANfAuq9oDye+wfIzZG7psZqDG3qJfcXHriuzdERERERlQgGvgR4dwbM7YCkW0Do/vzDw5p5wVitwrHQOzh3M0GrQyQiIiJ6Vgx8izB79mz4+vqiSZMmMHjGZoBv34fSHVxtzWXKg7CAs75ERESk5xj4FiEgIABBQUE4evQoyoW6A5XHoLVA9r0SZqNaKYvc1p2OQGwyS5sRERGR/mLgS4oqrQFrdyA9HgjekX9YdHKr72mHzOxcLD8artUhEhERET0LBr6kUBsBdZ9X9s+uzD8sOrnllTZbfCgU2SxtRkRERHqKgS/dU+9uusOlTUBGcv7h3g0qwMnKFLcS0rE1KFJ74yMiIiJ6Bgx86R4PP8CxOpCdBlzamH/YzNgIQ+6WNuMiNyIiItJXDHzpHpXq3qzvfekOwvDmlWGkVuFISBwu3ErUzviIiIiIngEDXyq8usPVnUBKbP5hdztzdK/rLvcX7OesLxEREekfBr5UkEsNoEIDIDcbCFpT4KXRLZVFbqtP3kRkYrqWBkhERET0dBj4UtEtjO9rZiH4V3ZA48oOyMzJxR+B17QzNiIiIqKnxMCXHlZngEj4BcIOAPFhBUqbBXT0lvtLDochLiVTi4MkIiIiejIMfOlhdhWVhhbC6eUFXmpfwwV1K9oiLSsH8/eHaGd8RERERE+BgS8Vzm+48nhqCZCbW3DWt713fmmzxPQsbY2QiIiI6Ikw8KXC1e4DmNoAd64rKQ/36VbHHd6u1khKz5bd3IiIiIj0AQNfKpypFVCnn7J/ammBl9RqFca3ry735+4NQVpmjjZGSERERPREGPjS49Mdzq8p0MJY6NPAA54OFohNycTyo/cWwBERERHpKga+VLRKzQAnbyAr5aGaviZGaoxrp8z6/h54DZnZ9/KAiYiIiHQRA196dAvjhkOV/ZNLHnp5oL8nXG3McCshHf+euFH24yMiIiJ6Agx8izB79mz4+vqiSZMmKNfqvwio1MoCt9irBV4yNzHCq22ryf1f91xFdg5nfYmIiEh3MfAtQkBAAIKCgnD06FGgvNf0rdZB2T+97KGXhzbzgoOlCUJjU7H+TETZj4+IiIiomBj40uP5DVMeTy0DcgtWcLA0NcYrbZRZ3282X0JKRrY2RkhERET0WAx86fFq9gLM7YDEG0BI4EMvj2ldVVZ4ELm+s3YFa2WIRERERI/DwJcez8QcqDvwXie3B4hc3496+8r9P/dew9XogqXPiIiIiHQBA196snSHC+uBtPiHXu7i64b2NV2QlaPBtHXnodFoyn6MRERERI/AwJeKx6MR4FIbyE4Hzv/70MsqlQof96kDUyM19l6JwZbzkVoZJhEREVFRGPhS8Wv65s36Hl8AFDKjW9XZCmPbVpX70/8LYitjIiIi0ikMfKn4GgwBTCyBW6eBSxsLPSWggzc87MxxMz4Nv+7mQjciIiLSHQx8qfisnIHmryv7Oz97qLRZXnmzqXcXus0JvIbQ2JSyHiURERFRoRj40pNpOVEpbRYVBJz7p9BTutd1RxsfZ2Rm5+LT9UFlPkQiIiKiwjDwpSdj4QC0mqTs7/ocyMkqcqGbiZEKOy5GYfXJG2U/TiIiIqIHMPClJ9dsHGDlAty5DpxcVOgp3q7WmNDBR+5/uPocrrG2LxEREelb4JuVlQVjY2OcO3eudEZEus/UCmj7jrK/5xsgK63Q0yZ09Ebzao5IzcxBwNKTSM9ilQciIiLSo8DXxMQEXl5eyMlhEFOu+Y8C7CoBSbeAo38WeoqRWoUfX/SDk5UpLtxKxGcbmO9LREREepbq8OGHH+KDDz5AXFxcyY+I9IOxGdB+irK/dyaQnljoaW625pg5uKHcX3woDBvO3CrLURIRERE9W+A7a9YsBAYGwsPDAzVr1kSjRo0KbFRO1H8RcPIB0uKAQ78UeVq7Gi4Y37663J/yzxmExaaW4SCJiIiIFMZ4Cv369Xuat5GhMTIGOn4IrBwFHJgFNBkLWDkVeurkLjVwJCQOx0LvYMKyE1g5rgXMjI3KfMhERERUfj1V4Pvxxx+X/EhIP9XuC7jXB26fAXZOB/r8UOhpxkZq/DTEDz1/2oszNxLw5caLmPZcnTIfLhEREZVfz1TO7Pjx41i8eLHcTp48WXKjIv2hVgPdvlD2j88HgrcXeaqHvQVmDGwg9xccuI6/Dl4vq1ESERERPV3gGxUVhY4dO6JJkyZ444035Obv749OnTohOjq65EdJuq1qG6W2r7B2ApB2p8hTO/u64f+61JD709adx9bzt8tqlERERFTOPVXgO3HiRCQlJeH8+fOysoPYRF3fxMREGQRTOdTpY8DJWylvtvFujd8iiPq+Q5pWQq4GeGP5SZwIKzpQJiIiItJq4Lt582b88ssvqF27dv4xX19fzJ49G5s2bSqxwZEeMbUE+v8GqNTA2ZXA+dVFnipaGk/vWxcdarogPSsXryw8hpCYlDIdLhEREZU/TxX45ubmykYWDxLHxGtUTnk2Btr8n7L/32QgKbLIU8Vit1lDG6G+px3iUjIxav4RxCRnlN1YiYiIqNx5qsBX5PdOmjQJERER+cdu3ryJt956S+b5GgIxey1msUUeMz2Btu8qVR5Ebd/1bwAaTZGnWpkZY+7IJqjkaIHQ2FSMWXgMaZnsCEhEREQ61sBC5PNWqVIF1atXl1vVqlXlsZ9//hmGICAgAEFBQTh69Ki2h6JfjE2VlAcjU+DyZuDk4kee7mJjhgWjm8Le0gSnw+Px6qJjSM9i8EtEREQ6EvhWqlQJJ06cwIYNG/Dmm2/KbePGjfKYp6dnyY+S9IubL9DhQ2V/8xQgJviRp1d3scbckY1haWqEvVdiMGbhUc78EhERkfYD36ysLBgbG8uKDl26dJEVHsTWuXPnkh8d6a+WEwGvlkBmMrB0EJAa98jT/Ss7YuHLTWFlaoT9wbF4ecFRpGZml9lwiYiIyPA9ceArFrB5eXkhJ4czcvQIaiNg0ELArhIQdxVYORLIyXrkW5pUccRfY5rC2swYB68x+CUiIiIdSHX48MMP8cEHH8j6vURFsnYFhiwHTK2BkEClvu8jFrvlzfyK4NfGzBiHrsVh1LyjSMlg8EtERERaXNwWGBgIDw8P1KxZE40aNSqwEeVzrws8/6eo3qu0ND4857FvaeTlkB/8Hrkeh5HzjiAyMb1MhktERESGy/hp3tSvX7+SHwkZrpo9gC6fAtumAls+UDq8+XR55Fv8vByw+JVmeGnuYRwLvYNO3+3Bm519MLJlFZgYPdXPa0RERFTOPXHgm52dLTtvvfzyy6zgQE+22C3mklLebOVo4JVtgOu9zn+FaVDJHivHtcR7/5zBqfB4fLbhAlYdv4FP+9ZF06qOZTZ0IiIiMgxPPHUmKjp8++23MgAmKjaVCuj1PVC5FZCZBCx+Hoi6+Ni31XS3wb+vt8TXz9eDg6UJLt5OwqDfDmLyilOISmL6AxEREZVB57Y9e/Y8zVupvDe3GLQIcK4JJN4E5nUDwg499m1qtQqDm3hh5/+1x9BmXjKG/vfkTXT/YS8OX4stk6ETERFROc3x7dGjB6ZMmYKzZ8/C398fVlZWBV5/7rnnSmp8ZGisnICXNwNLBwM3jgB/9QUGzgNq9XrsWx2sTPFF/3oY3LgSpvx7FhduJWLYn4cxvV9dDGnqVSbDJyIionIW+I4fP14+zpw586HXRP4va/zSI1k6AiPWAqtGK22NVwwHen8P+I8q1ttF7q9If3hn1Wn8d+YW3v/3LC7dTsL/etWGMRe+ERERURGeKkrIzc0tcmPQS8ViagkMXgL4vQRocoH1k4DdXz+2zm8eC1Mj/DzED//XpYZ8vuDAdYxecBQJqY9ukkFERETl1xMFvj179kRCQkL+86+++grx8fH5z2NjY+Hr61uyIyTDZWQMPPcz0PYd5fnuL4DV44DM1GK9XfzrwsROPpgzvBEsTIyw90oM+v2yHxdvJ5buuImIiMjwA98tW7YgIyMj//kXX3xRoHubqPRw6dKlkh0hGTaxUq3j/4Be3wEqNXBmOfBnJyDmSrEv0b1uBax6vQUq2lsgJCYFvX7ah0/XByEhjbO/RERE9JSBr+aBf4Z+8DnRU2vyCjByPWDtBkQFAb+3B879W+y31/Gww9oJrdDV1w05uRrM2x+CjjN2Y/mRMPmciIiIiCuBSHdUaQ28theo0gbITFYWv216D8jOLNbbna3N8PuIxvjr5aao7mKF2JRMWf2h3+z9OB56p9SHT0RERAYU+IqcSrE9eIyoxNi4AS+tAVpPVp4fngPM7aJ0fEuOLtYl2tZwweY328oqDzZmxjh7MwHP/3pAaXqRyKYXRERE5dUTlTMTqQ2jRo2CmZmZfJ6eno5x48bl1/G9P/+X6JkWvXX+GKjUDFj9GnDrFLA2QPyYpRyr2UOp++vsU+QlTIzUeKVNNfRtWBHfbrmIv4/dkE0vtgZF4o1O3hjVsipMjfkPHkREROXJE/3NP3LkSLi6usLOzk5uw4cPh4eHR/5z8dqIESNKb7RUvtTsDrx+AGj/PlChgfjRCwg/BGz/GJjVGPiz82MXwbnYmOGbgQ2wJqCVrP+bnJGNLzZeRPcfA7HncvFmkImIiKgczvjOnz+/9EZCVBi7ikD7KcqWcAO4tEnZQgKBG0eB39oC3b8EGo1UKkQUoWEle6x+vSVWnbiBbzZfxLXoFIycdwQdarpgTOtqaOXtxLQdIiIiA8d/6yX9YecJNB0LvPQv8OYZoGo7ICtVaX4hur+l3iutVxi1WoVBjSth59vtMaZ1VRipVdh1KRrD5x5G55l7sGB/CJLSWQKNiIjIUDHwJf1k66EsgusyHVCbABf/A35tCVzb/fi3mptgam9fbHurLUa0qAwrUyNcjU7BtPVBaP7FDvxvzVmExqaUydcgIiKissPAl/SXWg20egN4ZTvg5AMk3QL+6gts/R+Q/fiFltVcrPFp37o49EEnfNq3DrxdrZGSmYPFh8LQ9ftA/LbnKrJzcsvkqxAREVHpY+BL+s+jIfBaIND4ZeX5gZ+VhW/Rl4v1dhtzE4xoUUXOAC99pRlaVHNCRnYuvtx0Ef1/OYCgCLZAJiIiMgQMfMkwmFoCvb8HXlwKWDgCt88oC9+OzhV1+Ip1CbG4raW3M5aObYZvBtaHrblSA/i5WftkSbT0rJxS/xpERERUehj4kmER9X3HHwSqdQCy04ANk4FlQ4CUmGJfQgTAYhHc9snt0KOuO7JzNZi96yp6/rQX609HMP2BiIhITzHwJcNj4w4M/xfo9gVgZApc3gT80kKZ/S1G7m8eV1tz/DrcH3OGN5L1gEUJtInLTqLdt7vx595rrABBRESkZxj4kuEufGsRAIzdCbjUAlKilNnfn/yAI38AWcVvXdy9bgVsf6sdJnXygaOVKW7Gp+GzDRfQ8sud+HxDkHxOREREuo+BLxk293rAq3uAHt8ANhWAxJvAxreBnxoCh+YAWcULWu0sTfBWlxo4MKUjvhxQD9VdrJCUkY0/9oag3Te78P6/ZxAel1rqX4eIiIieHgNfMnwm5kCz14A3TgE9ZwC2FZXSZ5vfU2aATywCcou3cM3cxAhDmnph21vtMH9UE1kBQuQALzsSjg4zdjMAJiIi0mEMfKl8BcCi89sbJ5UKEHaVlAB43QSlAsTVncW+lOgC16GWK5a92hyrxrVAGx/nAgHwh2vOI6746cRERERUBhj4UvljbKbU/J14HOj6GWBmB0SeAxb1BxYPBKIuPNHlGldxxKIxzQoEwH8fv4kvThph9u5rLINGRESkIxj4UvkOgFtOBCadApq9DqiNgeBtSuvj1eOAyPNPHQA3reKALI0KP+wIRvcfArHncnSpfQ0iIiIqHga+RJaOQI+vgIAjQO3nAE0ucHqZEgCLGeCQwGI3wcgLgBe/3BgjfHLgamOG67GpGDnvCMYtOs4KEERERFrEwJcoj1N1YPAipQSabz9ApVZmgBf2AX5vD5xdBWQkF7sJhr+zBpvfaIVXWleFkVqFzedvo/N3ezBp+UmsOx2BhFTWASYiIipLxjBw8fHx6Ny5M7Kzs+U2adIkjB07VtvDIl1W0R8YtBCIuwYcnA2cXALcOgX8M0ZJh/BoBFRpDVRtA1RqrrRLLoKNuTH+19sXLzSuhKlrz+FISBzWnoqQmwiGG1d2QOfabuhaxw2VnazK9GsSERGVNwYf+NrY2CAwMBCWlpZISUlB3bp1MWDAADg5OWl7aKTrHKsBvb4D2r+vNL04vRSIDwNuHFG2fTMBtQng1Ryo00+ZJbZyLvRSNd1tsOLV5jgeegfbL0Rhx4VIXIlKxuGQOLl9uekC/q9rTbzerrqsGEFEREQlz+ADXyMjIxn0ChkZGdBoNHIjKjYRzHZ4X9nuhALX9wIhe5VH0RBDPIpt47tAtXZA3ecB7+6Fpj+I/F+xTelRC2GxqdhxMRJbzt/GoWtx+HbLJRy9HofvBzWEg5WpVr4qERGRIdN6jq+Yje3Tpw88PDxkYLBmzZqHzpk9ezaqVKkCc3NzNGvWDEeOHHnidIcGDRrA09MT77zzDpydC5+VI3osh8qA33BgwG/AW+eBiSeUkmgefoAmR6kFvDYAxt/XQqPrc4A7IUVeysvJEqNbVcWysc3xzfP1YWasxu5L0ej1016cCLtTpl+LiIioPNB64CvSD0RQKoLbwqxYsQKTJ0/Gxx9/jBMnTshzu3XrhqioqPxzGjZsKFMYHtwiIiLk6/b29jh9+jRCQkKwdOlSREZGltn3IwOmUikL4kRJtFd3K0Fwx/8Brr5Q5Wah0p0DMJ7TAlj/JpAY8YjLqDCoSSWsHt8KVZ2tEJGQjkFzDmLuvhD+6wQREZEhpTr06NFDbkWZOXOmXIw2evRo+XzOnDnYsGED5s2bhylTpshjp06dKtZnubm5ycB57969GDhwYKHniHQIseVJTEyUj1lZWXIrbXmfURafRSXM1gto8abccsKOImHte3BLPAMcnw/N6WXIbTwGuS0mKeXTCuHjYoF/Xmsmu75tOh+J6f8FYd+VKHzUqzY8HSzK/OsQfz8aCt5H/cd7aBiySvE+FveaKo0OTSmJma/Vq1ejX79+8nlmZqbMz121alX+MWHkyJEyfWHt2rWPvaaY3RXXEIvcEhIS0KpVKyxbtgz16tUr9Pxp06bhk08+eei4mCnOyxUmKi7H5EvwjVgJp5TL8nmW2hznPIchzKldke8RvyP33lZhTagaORoVTFQadKqYi04eGpgaleHgiYiI9ERqaiqGDh0qYz1bW1vdnfF9lJiYGOTk5MiZ2vuJ5xcvXizWNUJDQ/Hqq6/mL2qbOHFikUGv8P7778vUivtnfCtVqoSuXbs+8heyJH9i2bZtG7p06QITE5NS/zwq3fvo338CTIzfRPbVHTDa/TlMIs/CL2wuGjhmIqfbl4CxeaHv7wVgTGQyPt1wAYdC7mDzDSOcSzbHhz1roVMtF/lDIpU+/n40DLyP+o/30DBkleJ9zPsX+sfR6cC3JDRt2rTYqRCCmZmZ3B4kblBZ/mYr68+jUr6PtXsANbspJdB2fgb1qUVQR55VGmbYexX6Xl9PByx7tQU2nL2FzzdcwI34dLy+9BTa1nDBtD6+qOZiXebfp7zi70fDwPuo/3gPDYNJKdzH4l5P64vbHkVUXxDlyB5cjCaeu7u7a21cRE9FrQbavg0M/wewcFSaYvzWFgjeUeRbxMxu7/oe2D65Hca3rw5TIzUCL0ej+w97MWPLJaRl5pTpVyAiItJnOh34mpqawt/fHzt23AsMcnNz5fMWLVpodWxET827E/DaHqUEWtodYPHzwO6vgaz0It9iZWaMd7vXwpa32qJdDRdk5uRi1q5gdJ65B9uCWKWEiIhILwLf5ORkmYqQl44gSo6J/bCwMPlc5Nv+8ccfWLhwIS5cuIDXX39dlkDLq/JApJdEesPozUCjkWI5G7D7C+DHBsCBWUBGcpFvE+XOFoxugjnD/eFhZ46b8WkY+9cxjFlwVDbEICIiIh3O8T127Bg6dOiQ/zxvYZmo3LBgwQIMHjwY0dHR+Oijj3D79m1Zs3fz5s0PLXgraaKusNjE4jqiUmFiDjz3E1C5JbDjU6UL3NYPgb0zgGavA81eBSwcCk1/6F7XHW1rOOPnncH4c+817LgYhb1XYtCljhsGNa6E1t7OMGLrYyIiIt0KfNu3b//YIv0TJkyQW1kKCAiQm1glaGdnV6afTeVMgxeBOgOAMyuAfd8DcVeVGeADPwHNXwdaTwZMHy6lZ2lqjPe618LzjTzx8bpz2B8ciw1nbsmtgp05Bvp74gX/SrJDHBEREelAqgMRiR9BTYFGLwETjgID5wFudYHMZCDwW2B2M+DSpiLf6u1qjSWvNMd/E1tjZIvKsLMwwa2EdDkb3PbbXRgx7wjO3Igv069DRESkixj4EukStRFQ93lg3D5g0F+AXSUgIQxY9iKw9EXgTmiRb61b0Q6f9K2Lwx90ws9D/NDGx1l2VRZVIJ6btR8BS07gWnTR+cNERESGjoEvkS4SEatvXyDgMND6LUBtDFzepMz+Bs4Asu+11X6QuYkR+jTwwKIxzbDn7Q4Y4FdRXk7UA+7yfSDe//csIhOLriBBRERkqBj4EukyUyug8zRg3H6gShsgOw3YOV2pAHHwFyAz5ZFvF/m9Mwc3xMY32qBjLVfk5Gqw7EgY2n27Cz9sv4yMbC7eJCKi8oOBL5E+cK0FjFwPDPgTsPEAkm4BW94HfqgH7P0OSE945NtrV7DFvFFN8PdrLeBf2QHpWbn4YfsV9PppH45djyuzr0FERKRNDHyLIEqZ+fr6okmTJtoeCpFC5CvUfwGYdAro8yPgUAVIjVVKoX1fD9j5+SNrAAtNqzpi1bgWMgfY2doUwVHJGDjnIP635iwS07PK7KsQERFpAwPfIohSZkFBQTh69Ki2h0JUkLEZ4D8KmHAc6P874FwTyEgAAr8BZjcFLqwHHlEiUNQBFjnAog3yoMae8tjiQ2HoMnMPtpy/XYZfhIiIqGwx8CXSV0bGQIPBwPhDwAsLAfvKShOMFcOBpYOBO9cf+XZ7S1N8M7ABlo5thipOlohMzMBri47L6g/RSUUvniMiItJXDHyJ9J1aDdTppwTAbd4G1CbAlS3A7OZK/m925iPf3rK6Mza/2Ravt68uu70p1R/24N8TNx7bXIaIiEifMPAlMhSiu1unqcDr91WAEPm/33oDv3cA/nkF2PUlcOZv4OZxICutQAk00QVubUAr+FawRXxqFib/fRqjFxzFzfh75xEREekzrbcsJqIS5lJTqQAhAtyt/wNSooCIE8p2PzEzXKE+UKkZ4NlEPtatWBFrJ7TC74HX8OOOK9h9KRpdZ+7BlJ61MbyZl8wPJiIi0lcMfIkMkQhQRf6vSIGIDQZirwJxV+8+XgOiLwGpMcrMr9jyOPnAZPAiBHSojW513PHeP2dwPPQOpq45h90Xo/DNwPpwsjbT5jcjIiJ6agx8iQy9AoRbHWW7n8jdjQ8Dwo8A4YeVLfIcEHsFWDsBGLMN3q7WWPlaCyw4cB1fbb6IHRej0OPHvfh+cEO08nbW1jciIiJ6aszxLQLr+JLBzwg7VFbqAveaAYzbC7x5DjC1AW4eA07+JU9Tq1V4uXVVmfsrAuGopAwMn3sYX226iKycXG1/CyIioifCwLcIrONL5Y5dRaDDB8r+9mlASmyBzm/rJ7TGkKZecrJ4zp6rGPjrAYTGPrplMhERkS5h4EtE9zR9FXCrC6TdAXZ8UuAlC1MjfDmgHuYMbwQ7CxOcvpGAnj/uxcpj4Sx7RkREeoGBLxEVbIrRc4ayf+Iv4Maxh07pXrcCNk1qg2ZVHZGSmYN3Vp1BwNITiE99dL1gIiIibWPgS0QFVW4BNBgqVsABGyYDuTkPneJhb4GlY5vj3e41YaxWYePZ2+j+w14cCI7RypCJiIiKg4EvET2sy6eAuR1w6zRwbF6hp4gub+Pbe2P1+Fao5myF24npGPrnYXyx8QLSsx4OlomIiLSNgS8RPczaBeg4VdnfMR1Ijiry1HqedvjvjdYY2sxLPhfNLzrM2I2/j4UjJ5e5v0REpDsY+BJR4Rq/DFRoAGQkAFs+BHKLLl9maWqML/rXwx8jGqOivQVuJaTj3VVn0OPHQGwPiuTiNyIi0gkMfImocGojoNdMUfQXOPs3sKAnEBn0yLd08XXDjv9rhw971paVHy5HJuOVv45h8G+HcCLsTpkNnYiIqDAMfImoaJ6Ngd4zARNLIOwg8FsbYOtUICO5yLeYmxhhbNtqCHy3A8a1qw4zYzWOXI/D878ewI/bryCX6Q9ERKQlDHyLwM5tRPelPAQcAWr1BnKzgQM/AbObARfWK62PiyBmfKf0qIXd77THAL+K8tTvt1/G6AVHcSeFpc+IiKjsMfAtAju3Ed3HvhLw4hJgyArA3gtIvAGsGA4s6AVc3FBoybM8FewsMHNwQ8x4oYGc/d1zORq9f96H0+HxZfoViIiIGPgSUfHV7A6MPwy0eRtQmwCh+4HlQ4GfGwGHfgXSE4t860B/T6wJaIUqTpa4GZ+GF+YcxJLDoVz4RkREZYaBLxE9GVNLoNNUYNJpoPVbgLk9cOc6sHkKMNMX2Pw+kFb4bG7tCrZYN7E1uvq6ITMnFx+uPofxS04gNDalzL8GERGVPwx8iejp2FUEOk8DJgcp1R+cawCZScChX4D5PYGk24W+zdbcBL+95I/3e9SSTTA2nbuNzjP34JP15xHH3F8iIipFDHyJ6NmYWgFNxigpEMNWAdbuQNR5YF53ZSa4ECqVCq+1q471E1qjbQ0XZOVoMH//dbT7Zhdm7wpGWiY7vxERUckzLoVrElF5pFYDPl2AlzcDi/oBd0KAud2Al1YDbr6FvsXXwxZ/vdwUe69E48uNFxF0KxHfbrmEhQeuo25FO9hbmMDWwgT2liaySkQtd1s0r+YoA2ciIqInxcCXiEqWY1Xg5S3Aov5AVBAwvwcw/B+lJnAR2vi4oNVEZ6w7HSEDX7H4befFwtskT+3tizGtq5biFyAiIkPFwJeISp6NOzBqA7B0EHDjKLDwOaUcWvUORb5FrVahn19FdK/rjv3BMYhJzkB8ahYS0rIQn5aFm3fSZCm0zzcEwdvVGu1quJTpVyIiIv3HwJeISoelIzBirVLv9+pOYMkLQJdPgObjRZLvIzu/dart9tBxUfbs3VVnsPL4DUxYekKWRqvuYl3KX4KIiAwJF7cRUekufBuyHKj7PJCbBWz5AFj2IpAa98SXEnm9n/Wvi8aVHZCUno1XFh5DQmpWqQybiIgMEwPfIrBlMVEJMTYDnp8L9JwBGJkBlzcDv7YCQg8UPC83Fwg9qNQB/rMLEBL40KXMjI0w5yV/VLS3QEhMCgKWnkB2Tm7ZfRciItJrDHyLwJbFRCVIpDY0HQu8sh1w8gaSIpR2x3u+Aa7tATb8HzCzNjC/u1IH+MYR4J+xQHrCQ5dytjbDHyMaw9LUCPuCY/DZhgta+UpERKR/GPgSUdmpUB94dQ9Q/0VAkwvs+hz46zng6J9A8m3AzE55zaGq8nz7tCLLoH0/uKHcX3DgOubuC2HrYyIieiwGvkRUtsysgQG/Af1+BUytAQtHwO8lpfnFO8HKa8/9pJx7bJ6S/lCIbnXc8U63mnJ/+n9BeP7XAzh2/clzh4mIqPxgVQci0o6GQ4E6/QG1CWD0wB9FVdsqwfDJRcD6N4Bx+5Rc4QeMb18dubka/LL7Kk6ExWPgnIPo6uuGd7vXkiXPiIiI7scZXyLSHhOLh4PePF2nA1auQMxlYO93RVZ6mNjJB7vfaY8hTStBrQK2BkWi2w+B+HD1WdxKSCvd8RMRkV5h4EtEusnCAej5jbK/dyYQVfQiNjdbc3w5oD62vNkWnWu7ISdXgyWHw9Dm6114a8UpnI94eJEcERGVPwx8iUh3+fYDavRQagCve0MpefYIPm42+HNkY/z9Wgs0q+qI7FwNVp+8iV4/7cOwPw9h96UoLoIjIirHGPgSkW6XQev1HWBqo5Q4Oza3WG9rWtURK15rgXUTWqFPAw8YqVXYHxyLUfOPovsPe7HvSkypD52IiHQPA18i0m12FYHOHyv72z4Gdn4OxF0r1lvre9rj5yF+2PNOe4xpXRVWpka4FJmE4XMP483lJxGTnFG6YyciIp3CwJeIdF/jMUCVNkBWChD4DfCTHzCvB3BiEZCR9Ni3ezpYYmpvXxx4vxNGtqgsJ5LXnIpAxxm7sfRwmKwMQUREho+BLxHpPrUaGP6P0vq4eieRAwGEHQDWTQBm1AC2fAjkZD/2MnYWJvikb12sGd8KdTxskZiejQ9Wn8ULvx3kAjgionKAgS8R6QdRx7feQOClf4G3zgOdPgacfICsVODgLGD5ECAzpViXalDJHmsDWuF/vWrL1sfHQ+/IBXADfz2AlcfCkZaZU+pfh4iIyh4D3yLMnj0bvr6+aNKkibaHQkSF5f22mQxMOAoMWgQYWwBXtgILegPJ0cW6hLGRGq+0qYbtk9uhd/0KcgHcsdA7eGfVGTT9fDv+t+YszkckgkUgiIgMBzu3FSEgIEBuiYmJsLOz0/ZwiKgwIlnX9znApgKwdBAQcQKY20VJi3CqXqxLeNhbYNbQRohKTMfK4zew4mg4wuJSsfhQmNxMVEaYcTEQLrbmcLE2g4uNGbwcLfFiEy/YWZqU+lckIqKSw8CXiPRfpSbAmG3A4gHAnRBgbldg6N+Ap3+xL+Fqa46ADt54vV11HLwWi2VHwrDl/G1k5QA34tPldr/5+6/j64H10a6GSyl8ISIiKg0MfInIMDh7K8Hv0heAW6eBhb2Bxi8Dnk2UTaRHFINarUIrb2e5Jadl4O91m1G3SUvcSctBVFIGopMysP50BEJiUjBy3hEMa+aFD3rWhpUZ/zglItJ1/JOaiAyHjRswagPw90jg6g5l0Vv+ax6AZ2OgcivAfyRgYvHYy5kZq+FkDvhVsoeJyb20BjEr/PXmi1hw4LpsjbwvOAYzXmiAJlUcS+ubERFRCeDiNiIyLGY2wNAVSukzUf/XvT6gMgKSIoAL64DN7wFzWgNhh576IyxMjTDtuTpY8kozeNiZIzQ2FYN+O4gPV5/FrotRSErPKtGvREREJYMzvkRkeIxMlNJnYhNEmbOIU0rb48O/AbHBwLzuQLNxQKepgKnVU32MSIfY/FZbfLo+CKuO35Czv2JTq4C6Fe3QrKojmlV1QktvJ1ia8o9bIiJt44wvERk+EdhWaQW0fgsYfwjwGw5AAxz+Ffi1FXB931Nf2tbcRKY5/PVyUwxq7InKTpYQjeDO3EjAH3tD8Mpfx9Dqq534eccVJKRyJpiISJs4BUFE5YuFPdB3NuDbH1j/hlIFYkEvwLszYOGo5P6KQNnEAmpjK1ilF282uG0NF7kJtxLScPhaHA6HxCLwcgxuxqfhu22X8VvgNQxr7oUxravC1ca8lL8oERE9iIEvEZVPPp2V2d9tU4HjC4Dg7Q+dYgSgg8oEOKoBWoxT6gYXQwU7C/Tzqyi37JxcbDx3G7/sCsbF20n4bc81WQptcONKmNjJmwEwEVEZYuBLROWXuS3Q50fAbwRw+zSQmQpkpQFZKXI/99ZpGIUfArZOAa7tUGaKReWIJyA6xD3XwAN96lfAzotRmLUrGCfD4rHoUCjWnLqJd7vXwtCmXrJzHBERlS4GvkREotFFIc0ucjIzcO6vyah3ayVUwduAX1sAz80CavV84o9QqVToVNsNHWu54tC1OHy56YLMA5665pxcGPd5v7pyQRwREZUeLm4jIiqKSo0Ql67IHrMDcKsHpMYCy4cA6ycplSKe5pIqFVpUd8Lq8a3wyXN1YG1mjNPh8Xhu1j5M/y8IyRnZJf41iIhIwcCXiOhxXGoBY3cALSeK0FXJCf6jIxB18akvKVIbRrasgh3/1w6961eQlSDm7gtBxxm78evuq0hIYwUIIqKSxsCXiKg4jM2Arp8BI9YC1m5A9EXgjw7AqWVFv0ejAWKvKnnDRXCzNcesoY2wYHQTeDlayrbIoitcyy93yPrA4XGppfN9iIjKIQa+RERPolo7YNw+oFp7ICsVWDMOWBugLIzLEx8O7J0J/NIc+LkR8Ht7IDnqkZdtX9MV2ya3lTWBa7rZICUzB/P2h6D9jN2YsPQEjoTEIVdMCxMR0VPj4jYioidl7QoM/xcInAHs/hI4uRi4eQLwHwUErQNCH2iIIWaHF/QGRq5/ZFUIM2MjDPT3xPONKiLwSgz+CLyGfcEx+O/MLbm52ZqhR90K6FmvAhpXdoCalSCIiJ4IA18ioqehNgLavwd4NQf+eQWICgI2vXvv9SptgPqDAPd6wPJhQMwlYGFe8Ov+2AVw7Wq4yO18RAIW7L+OzeduIzIxAwsOXJebq40Igt3RrqYLmlZ1kovkiIjo0fgnZRFmz54tt5ycHG0PhYj0IfVBVHpIvAnUHQDUHQjYV7p3zqj/gAV9gJjLysyveP6Y4DdPHQ87fPtCA3zWvy72XYnBhrO3sC0oUuYCLzwYKjdjtQoNKtmjVXUntPR2hp+XvZw9JiKighj4FiEgIEBuiYmJsLNjbU0iegSRvjB0edGvO1ZTgt2FfYDYK0qL5JH/AbYViv0RIpAVdYDFlpGdg/3BMdh6PhL7r8YgPC4Nx0PvyO2nncGwNDVCGx9ndPF1R6darnCwMi2Z70lEpOcY+BIRlQXHqndnfnsDscFK2sNLawrODD9BENyxlmiGoeQLi8oPB67GYH9wLA5cjUVMcga2nI+Um0gDblzFEV193dC5thuqOFuVwpcjItIPDHyJiMqKQ5V7aQ8i+P2tDdD/d6BG12e6bCVHSwx29MLgJl7QaDQ4H5GIrUGRMiXiwq1EWRFCbJ9tuIAqTpaygkT7mi5oXs0J5iZMiSCi8oOBLxFRWQe/ozcCf78ERJwElr4AtJ4MdPgQMHr2P5LFwjjR+lhsk7vUkLPB2y8oQfDR63G4Hpuav0DO3EQtg98GnvbwcbOGj6sNqjhbMj+YiAwWA18iorIm0hte3gJs/R9w5Hdg30wg/DDw/Nwnyvst7mzw6FZV5SbaIYvc4N2XorD7UjRuJaTLR7Hd31GuspMlarvbom0NZ3So6QpXW/MSHRMRkbYw8CUi0lYnuJ7fKuXQ1r0BhO5XUh86fwKo1EBK9N0tBkiLU8qjNR8PqJ++75AoedatjrvcRErEpcgkWSnicmQSrkQlIzgyGUkZ2bgWnSI3UUFCqFfRDh1ruaJTbVfU9bBj/WAi0lsMfImItKnu84B7A+DvEUDUeWDt+MLPu7wZCNkDDPgDsLAvkZSIWu62cssjgmFRK/hKVBJOhMZj56UonA6Px9mbCXL7cccVGTx7u4q0CGuZHqHs26CivQUDYiLSeQx8iYi0zdkbGLsD2DFd6fpm6QRYudzdnIHcbKVL3JWtSvvjF5cAbnVKfBgiGHa3M5dbGx8XTOrsg6gkJR1i18UoBF6OlukSp8Lj5XY/W3NjWUu4vqedzBkW+25MkSAiHcPAl4hIF5hYAN2/KPp17y7AipeAOyHAn52B534G6g0s9WG52phjUONKcsvKycX1mBSZFnElMlnODAdHJcu0iMT0bOy9EiO3PGIWeHSrKhjevDKrRxCRTmDgS0SkDzwaAq/tAVa9DFzbBfwzRqkK0ekjJV+4DJgYqeHjZiM31Lt3PDM7V+YJn74RL1MjztxIkM9vxqfJEmpz94VgUicfDPT3hLHR0+coExE9Kwa+RET6wtIRGP4PsHM6sO974OAs4MRfQI3ugO9zgHdnZea4jJkaq/NLqA1rVlkeS8nIxn9nIvDD9iuyesSUf8/it8BrssRar3oVkKvRID4tC/GpmYhLUR6ruSg5w0REpYWBLxGRPlEbAZ2nAR5+wKYpQFIEcPZvZTOxUpph1OkP1OxVInWBn5aVmbFsqNG3YUUsORyG2buCERKTgonLTuK9f84gNTOn0PdVc7GSVSdEpzmRK/zggjmRbhGZmA4rU2O2YiaiJ8bAl4hIH/n2BWr1AW4cBYLWAhfWAQnhwPnVyuZYHWj7NlBvkFYDYJHbO6Z1VQxuUglz94bgj73X5AK5PHYWJnC0MoWVmREu3U6S+cK/7r4qNzdbM7T2dkFqZjYiEtJxOyENUUkZ0GiUesNvdvLB6+2rM32CiIqNgS8Rkb4SNX29milbt8+BiBNKECzSH+KuAmteB/Z8DbR5G2jwImBkorWhijJookrEmDZVZQDrYGkqg977g9ak9CzsuhSNredvy0oSorTaPyduPHQtY7UK2bkafLftMnZdisL3gxuispNVGX8jItJHDHyJiAyBSgVU9Fe2tu8AR+cCB34G7lwH1k0AAr8BqncEstKAzBQgM1l5zE5X6gj7dAGqdwDM7Up1mEodYJtCX7MxN8FzDTzklpGdgwNXY3Ey9A7sLU3hYW8BD3tzVLCzgJOVKdaevomP1pzHibB49PxxLz5+rg5e8PeUJdmIiIrCwJeIyNCY2QCt3wSajgWOzQP2/wTEhwHHFxR+/u2zwKnFgMpI6SQnFslVbqWkSOTmKJvm7qOptdJyWdQYfpogMzsTiLkM5GYBOXc3sS+uXbERYOGgfAVjI9kuWWyF6e/nicaVHfF/f5/GketxeHfVGey8EIXJXWvA2dpMziaLdAgiovsx8CUiMlSmVkDLiUDjMcrit6TbyjG5WSuPwvX9QPA2JSAVrZPF9jjG5oCdJ2DvpWx+IwBP/0e/J+4a8FdfJQgvjGjc0ecnoHbvYn29So6WWPZqc/wWeBXfb7uMzedvy+3+phpiAZyzlSlqm6rQQyQHE1G5xsCXiMjQmVoC/qOKfr1WLwBfKGkRV7YBwduB2+eUGV1RRULMBOc9ZiQCiRFKikRssLIJp5YC/X8D6g4o/DNirwILeitVKETQLVIqRM6x2kR5TE8AEm8CK4YBfi8B3b8CzB5f2kzM6o5v7422Pi74aO052Vgj6e7iOdFUQ2yhsak4DiNc/+sEvhxQXwbMRFQ+MfAlIiKFQxUlPUJsj0tXEEGqqCIRH64sqLuyBVg1WgmKW04oeH7MFSXoTb4NuNQCRq4HrB9IYcjOAHZ9rqRlnFwEXN8HDPgdqNS0WEMXNYT/Hd8qv+RZwt0awXdSs7D/SjRm7byCfcGx6Pp9IP6vaw2MalmF1SCIyiEGvkRE9GSMTQHHqsomiIoRm6cAR34Htn4IJNxQqkyIWeLoS8DCPkByJODqC4xYB1i7FHJNM6DLp4BPV2D1OKU187xuykK9tu8+UUk20WFO5PmKTWhY0QZWsRexPcEFh0PuyG5y605H4IOetWFpaoSk9Gy5iTJryelZqO5qjdbezlwoR2SAGPgSEdGzEQFuj28Au0rAtqnA4V+VGeE2k4Elg4CUKMCtLjBiLWDl/OhrVWkNjNsHbHxHyUsW5dhEDvIL8x+eJX4CrhbAogGNsfr0bRn4irbKL/5+qMjz61a0xYQOPrKRxoNNNASNRoOgW4mIiE+XQbKFqdFTj42Iyg4DXyIienZidrTVG4Cth1I/WDTUEJvgXg94SQS9TsW7loU98PwfQI1uwPpJQOg+4Ld2wOBFgGfjZxiiSnaTE5Uipm+4gL1XomFpYiTLqFmbG8tSa2bGauwLjsG5m4kYt/g4arhZI6CDN3rX95ApFAevxmL7hUjsvBglWzELDpYmeKl5ZbzUogpcbJRZ5tKQnZOLTeduo6KDBRp5KdUviOjJMPAlIqKSU28gYO0GLB8GZCQAFRoAL60BLB2f7loiaF4xXKk4Mb+HMrPcePQzDdHV1hw/D/Er8vW4lEzM3x+CBfuv43JkMiYtP4VvNl+Sx9Oy7rVatjAxgr2liQyAf9oZjDmB1zDAryLGtnBHdXvjp/vORbgcmYR3Vp7G6RsJ8rl/ZQe82rYautQufEaaiArHwJeIiEpW1TbAq7uAy1uAhkOVGdyn5VITeGWHMot88T/gvzeVDnU9vgVMzFEaRAvl/+taE6+0qYa/DlzH3P0huBmfJl9ztzVHp9qu6FzbDS2qO8l8YtFp7rfAazgVfgfZJxbD9cxfgCoN4TYNkFytN5yaDoJrxSpPPcsrrv3j9ivIzMmVs9KZ2bk4HnoHry06jmrOVnKcAxpVlO2hDUFaZo6ceWdAT6WBgW8RZs+eLbecnHs/3RMRUTE5VQdajC+Za5nbAoMXA/tmAjumKy2ZbxxTZoRFNzrReU60by5hognGxE4+GN26KvZdiYangyXqeNg+tOitR70K6F7VGPF/B8AhbEv+8UpJp4HTp5F76gucUtdCkEMnHLHvjsh0E1l1Im8TgWydirZoXNkBjas4ykcnazNcuJWId1adlmkXQsdarviifz2IeHDBgetYfCgU12JS8MHqs/hu6yX0qOeOTiIgr+akt0Hw4WuxGDHviPyus4c2YvBLJY6BbxECAgLklpiYCDu70m3hSUREjyGCzTb/B1RoCPwzBogKAnZ8qmyi8UW19koQ7NUCcKhaooGwmGXtXrdC0Sdc2gzVuglwSImWdYnjm72DvebtoLq4AVWjtqFOzgU01FxAw7gLqBWzGS9kfowcFAxMT4bFy+2PvSHyeXUnc9yMT0V6jloG4B/38UV/v4r5Qfe73WthfAdvrDgajnn7lBnpxYfC5CYqVYgFd2JWumNt1/zqFrouMT0Lk/8+jYxsJZf5l93BmNDRR9vDIgPDwJeIiPSHdydg/GFl4dzVXUBIIJAaC5z7R9kE0SBDlE4T+cHudaFy9oVKtFwuCaLecHqi0nBD5DAfXwicWKi85lJb1h62r1AffcTzds0BTEdaTCiiD6+E+8nv0Sg7GGv9TyOm/muwtzSVQa2oEHEqPB5Hr9/BsetxuBEVgxlJH8HTOBrzvaZh1JChMi+5sIB8TOuqGNmiMvZeicG2C5HYcSESkYkZ2BoUKTdTIzWGNK0kF+gVdg1d8sm6IBnAi457ovHId9suo0Ele7TxKaT8HdFTYuBLRET6xcbtXqONnCwl7eHqTuDaLuD2WSAzGbhxRNnu/kXXzdgGavUeoN7zQOVWSgm2B6XGAZHnlQ52or2z6DInHkVTDlGHOO2O0rHuISqgRQDQcWqheccWzpXh1ettoIIbsG4C6l6aBXQYAjh7559TzcUaAxp5ijppyFz5CkyDlI5470RNgSrcEajTr8hfDtGIo0MtV7lp+tXF+YhEbAuKlJsoubbwYChWHAvHiBZVMK5ddZnDrGs2n7uNf07ckGkc80Y1warjN7D8aDjeWHYS/73RBq5WDFeoZPD/JCIi0l+i3XHlFsrW8UMgJ1tpoxx5Drh9RrZe1tw8DrP0eODEAmWzcgV8nwM8myjVIkR7ZnG+qD1cXGa2ymZXUQl4xYK+x/EbDpz/VwnS1wYAozc9nJJxfD5Mg1Yp7aG9mkMVuh9YOQpI/hpo9tpjP0KkQogudmJ7q0sNHAiOwYytl3AiLB6/B17DkkOheLl1VXSv6w61SiUzSFRQyYDTzNgIlRwtyrxxR1RSusxTFkRgLvKcxfjPRSTI/ObxS05gyctPX8aO6H4MfImIyHCIDm+utZRNLH4T2QnpqTi6ciaaW0dAfWmD0lDj6J/KVljbZidvwKaCstnefRQl2kQusfndgLewGePHEQFln5+AX5oD4YeUTnfNx917PeIksOk9Zb/zx0CLCUojj2NzgU3vKoF5p2lPlL/c0tsZ/1R3wu7L0XIBnAgkf94ZLLfCNPKyx8d96sgUg7Ig0jym/HNWlorzrWCLNzvXkMfF4rxfh/mj98/7cDo8Hl9uuoSm+rlej3QMA18iIjJsRiaItq2HnJ7vQf3cj8C1PcD51UDcVcC5hpILLDrLudVRAtvSZF9Jac28YTKw4xOgRlfAsZqSRvH3CCAnE6jZC2j5hhIo9/pOaQqyczqw/0cg8RbQd7bSNrqYxAyuaNrRvoYLtpyPxK97riIiPk1kVYjQE7kaJQAVLZvFzHDf2fsx0N8T73avCVeb4uUF5+ZqcCUqGadvxMsmHqL6RXHeK9IZRDMQkYv8/eCGMDW+F9RXcrTED4Mb4uWFR7HkSDhU3ir0fNTFsjOB1BhALDIUaSuihnQJ1lImw8DAl4iIyldqhE9nZdMW/9FK4H19L7DuDaWV8+rXgfgwwL4y0O8XJegVxGPbt5VZ5/VvKG2c40OBfr8qJeOegAiARYqD2AoTmZiOrzdfxL8nbsoc201nbynl3FpVkWkQhQW6h67Fyu1wSBzUKdEYabwFf+a0wGVNpfwAWGw13W3hYm0GJ2tTOFiKzUQuZJv+X5C83jvdaqKmu81DYxJ5yxM7+uCnHVew4poavsdvYIC/171ybbu+BM6tUoJdseDwfhX9gbE78VTEDyLzeyl1pEW7bDIYDHyJiIjKkkhVeO4n4NdWSvC78DmlLbORGTDor8IbfvgNUxb1/T0SCD8MzGkNdP4EaPJKiZVuc7M1x8xBDWX75Wnrg2SKwVebLmLOnquyoURWjgZZ2bmykYbYlBljhQMSsdTsC9RQheMFkwNol/4NopOA3Zei5VYYMcsrrtOsqqOsTlGUSZ18cDI0DnuDY/HBmiDM3B6Ml5pXwchqibDf81XBk0VutJUzkBID3DwORF0AXGs/+S/GkT+BqPPK1u7dp7sG6aSSr/hNREREjybSGzp9pOyLoFfo8TXg0bDo93h3Bl4/AFRpA2SlApveAf56DrgTWqJD8/NywOrXW2LmoAZwtTFDfGqWLJEm8nCTMrJlnV0R9JqbqNHGxxkfdvLAvoqzZNAruGuicLZbMP55vSWm962DF5tUQpMqDqjuYiVnevMms0XQK8q5zXihwSMbVRipVfhlaEP0rZwDd1szxCRn4vvtl3Fwwfvy9ZiKnbG360as7boP8zoexU9+G3DZrqXy5rOrnvwXICsNODzn3vPTy5/8GqSzOONLRESkDU1fA86vURa61X8R8B/1+Pc4VAZGrFMWvG37SJkx/rUl0HU60GhUic3+ikBUlFfrUbcCLkcmyeBT5N+KFs0mRio5WyvqEJvmpACL+gOx5wBLZ6DJGGDP1zA98AP8/UfAv/LDrZpzcjWIT82UgbToUFec8moitaGjhwZfjm6D7ZdisWXXbnS7o5SrG3qtCy5fjQcgNsUVdQP8bLoXcYeXwqrt+zAzKTrcCYpIxLWYZDSt6qjkJZ9aouQKq02A3CzgzN/KDylPs6CRdA4DXyIiIm0QQeqQZUDwDqW8WnHLiIn3iRrGopnHmgAg7ADw31vAoTlAy4lA/UGAccl0a7MwNSq6woOYGV02BLhxFDC3V3KVReOQ4O1KmoFYkNd31kNvE0G0CHjF9qRE4P1cAw/0Cd4JVbwGJ63aAHa10cjMGDbmJrA2N5YNMLLTeyHl0h9wzIzA5J8XYOKIIajqbFXgWiExKZix5RI2nL2Vf8yvojXmp8yE+MaaTh9Btfc7pZ6zaJRSvcMTj5d0DwNfIiIibRFVB+q/8PTpEqM2AEd+A3Z+DsRckg0ysPMzpUyaWET3YL6wyFHISAJMLJXSb8/SwW7FcGXG2dQGeOlf2SVP6v4VMLcLcHKxEqCL6golKfYqVHe79PkN/xxbi7j+7fldYRW6HnXjtqH3T574rH9d9PfzRFRiOn7ccUVWlBCzz+LnDW8Xa7lYr+KtbbA3vYk4jTX6766GL23ao2X6GmQcXwozBr4GgYEvERGRvhKzv81fBxoOBY4vAA79CiTdArZPAwK/U6pXZCQrVQ/Egi/xmJOhLKQTtY5lGbe6StAqHi0cip55FjO8t84os7kX1iszzSKAHrZSqaCQp1JToO5ApdrC5g+AUf8Vfza7OMQsrCYX8On2yKDaveVwIHQ9+psexmepw/HWitOyWsWJ0HikZSktrDvWcpUVJWpXsEVUYhqM//wMSASWarojNEmFb5P9sNpsDXLOr8XAn15AYx8vmdfcvJqTnLnWNSkZ2dgXHIPW3s6wMmOIVxj+qhAREek7czug1SSg2etKwHngZyAqSCmbVhgR/N46rWz3M7YArF2VTXS4s3ZRgkzRXCMyCNAoAaMkgmeRqiG65j2o8zTg4n/Kwj0RJItUjieZTS4qVUOUcstbbCaqLTxK9Y4yBcMh/Q5mNEnC28fssD84Nr9Rx5QetWVebx7X6ENAYpAM5sdO/Ar1bwN7L1fBjRO/wTM3Al63d2BORFtZ5aKWuw2m9vZFK29n6IqM7By8NPewrMUsSsm907Umnvf31MkAXZsY+BIRERkK0dhCzP42GAJc3aG0YxblvcTCMyuXu/tOSve6yPP32jXfPqsEldlpyqPYCiOCYc/GgEcjoHbvost8iUYdIt848Ftg21SgRrfH5x2LNIzDvwHbPwY8/JR6xiKd4z7qAz8qwXe1Dso4Hvdr4dsXOLEQA0wOwvPVT7D0cCh61KuArr5uD7dm3v+D8thoBMxsXdDWFmhbwwWweRnY9RmmeJwCXIdge1AkLt5OwrA/D6OLrxs+7FkbVR7IHy5rogHJ1DXnZNArRCdl4N1/zmD+geuY2qu27OBHCga+REREhkYmrnZWtsKYWStBZe0+945lpgDJkUCySIuIApLvbrnZSkpBxUaAbcXipy20ehM4sQi4c10pDyZmpIsimk+snQBcWKc8DzsI/Noa6P4F0GikPGSeGQv1hWXFm+3NI9pWn1gIBK1D057foWlVv8LPEzPa13YrdYCbjy/4mlgsuOszuMYcxsyXnBHf2xc/bL+CRYdCsS0oErsvRWF0q6qY0NEbtuYm0IYFB67j72M3ICZ3/xjRWC7cE3nMF24lYuifh9G5tive71kb1V2sUd4x8CUiIiLA1EoJhh+YZX1qIrju/DGw5nVgz7eAQxXAuwtgalnwvIhTwMpRwJ0QpYRYh/eB4J1KmsT6ScClTUCPmfCJ3ACVKC9WuTVQ+W6d3sep3AqwdgeSbysz4DV7FH7e/p+Ux7rPKyXj7ieei88U4znzN+zbTMa05+pgeHMvTP/vAvZcjsbvgdew7HAY6lS0RU03G9Rwt5HpED5uNrAwMUJscqachY1JzpCPcamZskayCESruVjJihRPa39wDD7bcEHuT+/kgk7xq4Bmo2Q5OtHxTgTo2y9EyUYiw5p5YVLnGsUqIWeoGPgSERFR6RD1iY/8rsyo/j1CWQzn0wWo/ZyS/iBq5G5+X8k5tvMCXlgAePoDrUR5ttnAjk+By5thfOMoKqclKtds907xP1/U3q07ADj0i9LMorDAN+4aELRG2S9qVrrBi0rgK/KLW78lZ729XW2w8OWm2HUpCp/9F4Sr0Sk4dC1Obk+qgXUCXjfbDL/sU8g1MkO2sTWyTayRY2IFjYkNcj0bo1rnsbKc2/1CY1MwfskJWZ1iQEMPDL0xVam0kRgBx26f3w3QK+PLjRew42IUFh4MlS2pAzp6Y1TLKvdaP5cjDHyJiIio9KpODF2p5M+KNIb4MCBorbKpjZU0CqFGD6D/r0pVibz3iRxhsUDt39egijwLEaLlVmwCddV2TzYGUWFCBL6XNirpHGJmO09OllIKTizgE7PReSXZHiRyhTe+rZSME0G8SPu4q0NNV7TxdsaFW0m4FJkkG35cuq083kpIl+eIBWbO1qZwtjaTC88cLE1xOyEdqqhzeCHjX/TJOgjj7Nyiv0P433j1UDKc63dFn/oeclGeqEwx9q9jSEjLkrWWv/KLgWrZXuX8U0uVphvGZvB2tcbcUU3kzPDnGy4g6FaibEW96GAo3utRC33qV3g439mAMfAlIiKi0iMqQ3T7HOj6GXDrlMy3lYFv3FUlp7bLJ0CLCYXnDrvVAcbuQM7OL5ByYhUsu3wO9ZMGaSJIdaiqpFKItAmR9yuIxX0iDSOvsoWYyS2KuS1Qq7dSMUPM+t4X+ArGRmrU87ST2/1EUJqdkysD3fy2zGIRX+gB5YeB7K2QEb0YjnMLHHUegFSNEVQZyVBnJkGdnYLqScdRP+0wJufMR6/DPlh6OEymSYgA+nJkstz/fbgfTJffN5udFqdU1RCpG3eJChT/TWyNf0/elI07bsan4Y1lJ7Fgfwg+7VsXdSsWHPszEd9RlL97MK1FBzDwJSIiotInAlZRrUFsYjYy+qKS0+vs/ej3GZsht8P/sCutEXo+EHAW+3NFALh3BiAaX4jZ230/yNbKsiWxmGXuOQOo0urR1xGVMkTgKzYRxIuqEY9hZ3Ff7m5ONnBhLXBgFhBx4u7Y1Mp4Wk2Cm4cfehd2kbQ70PzUCLXSwvFd1eP46FZLRCVlyE20kf7tJX+43dii/FBhaq0sxjs2Dzi+sEDgK6jVKgz090SvehXw595r+HXPVVkJos+sfTL/9+2uNWUr6uJITM9CWGwqwuNS4eVkiToe9wXOIt1CNDhpMhboNBW6hIEvERERlS0RjBZVCq00iFleEfhe2Qb82eneLG/NXkDv7wEbt8dfo1p7wNpNqXxxcf1DQWWR0hOBk4uUltIJYfdqIDccArR8A3Cq/uj3WzhA1fFDYMP/od+dBeg5eSL23syRucWdarvBr6IN8Mtnyrli5txvGHBsPhCyR8lfLmSxooWpESZ28sELjSvhi40XsO50BBYfCsPGs7fxbreaGNS4kgySRe6wCGxlCsftJARHJyM0NhVhcamIS8kscM2mVRzxSpuq6FzbDWoReItKHWLmWccw8CUiIiLDJoJs1zpA1Hkl6DW3B3p+C9R7ofjl2USLZ3H+wVnAqjHApc1A+ylFB66iNrLItRWtmzPuLswTNZTFLGiTV5QUkOIS7adFMBt5DqaBX6BT75ky6JXEzG5sMGDhCLQIUNIyvDsBwduBE38pzUSK4G5njp+G+OHFppXw8drzsm3zlH/PyvJoYiHdlagkpGcVnXvsZGUKD3sLWTbtyPU4uTVyzMLKtPVKBkfjl6FrGPgSERGR4RNB4drxQM2ed2d53Z/8Gu3eAxLClRzls38rqRN+w5W6wnaeSt1jUalC5AFHnr33PucayufXHwyYWDz554rqFD2+Bhb0Ao7PBxqPBtzrKXm0ImVDaPu2EvQKovaxCHxPLgE6fAgYPbpcWsvqztg4qQ0WHrguaxSLBh15zIzV8HGzRg03G/i42qCKk6VMbajsZAXru22RxUI9ESyLBiHNEtbByCQbZ+CDPefNMfEpfplLEwNfIiIiMnwiBaBOv4JVHZ6UCCwH/aXUHt75GRC8TWmQcXoZULExEH74XltnI1OgRnfA7yWlkYioVPEsqrQGfPsppdc2TQFG/Qcc/RNIvAnYegKNx9w7V5RtE132RCOSy5sLNiopgpjhfaVNNfRp4IGtQZFwsTZFTXdbeDlaPrbtsZg5ntKjFiZ2qIbcH94E0oGFmZ1gfCcNuoaBLxEREZUPzxL03s+jITB8FRB2CNgxXanxG3ZAeU0EwCJ/t84AwNIRJarrdCWQFZ8n8ob3zlSOi5QLE/N754kZXtG6WlSOOL6gWIFvHjdbc7zU/IEmHsVkFR4IpEdAY26Hrv3GwafiE6RzlBEGvkRERERPw6u5MvMaEqiURxMzuy41Su/z7L2UVtB7vgLWvSHqhilpFKLixIMajVAC3+AdSv1k8d7iELWNb54AKvorec1PQixqE2sXGwxFt4ZVoYuecd6diIiIqBwTi+OqtQNajC/doDeP6C4nUhtE0Ct0/F/hAapYdFe1rXKeWGBXHJkpwOLngXldgRXDlBJsxZVwE7i8SdkXOcg6ioEvERERkb4QTSG6f6HsezZR2j8XRSxyE0Tg+7ggNiMJWDxQKYMmiJSKze8pzSiKQ1SQEB3wKrcGXGpCVzHwJSIiItInounFa4HAsFWPLscmcntFmTOxAE5UeShKWjywqL+Sp2xmC7R/X0xlK4vnRPm2xxFBtVjkp+OzvQIDXyIiIiJ9U6EBYGH/6HOMze7l/+YFpg9KjQP+6gvcOKrUNx6xVlksJ9pMC1v/B5xfg0cSs8NJtwBL5ydaSKcNDHyJiIiIDJX/3XSHSxuBn/2BtQFK6kPsVSA5GljYR2l3LJpriIV6eW2hm48Hmr6q7P/7KhB2+LGL2mRNYxFs6zBWdSAiIiIyVCLftuFw4NQSpcOb2PIWu6mNgdxspRXziHWAa6177xMpFN2/AuLDlUVry4cAY7Y93KkuLgS4ukPZ9x8FXcfAl4iIiMiQ9ZsNdPsMCD8KhB1U6g/fPA7kZAA2HsDI9YCzd+Ed4wbOVTrGRZxUZodFUw7ZAlpsvkqdYKF6J8BRN0uY3Y+BLxEREZGhs3AAanRVNiE7A4g8BzhUfXSjDdH0Y8gKYG5npR7wsbkPnHB3cV3jl6EPyk2Ob2pqKipXroy3335b20MhIiIi0i5jM6VJRXG6y9m4Aa/tBfr/pjTQELO++Q0xNErwLI7pgXIz4/v555+jefPm2h4GERERkf6xsAcavFjwWEYyEHsFsPN68i5vWlIuZnyvXLmCixcvokePHtoeChEREZFhMLMGPPwAKyfoC60HvoGBgejTpw88PDygUqmwZs3DteJmz56NKlWqwNzcHM2aNcORI0ee6DNEesOXX35ZgqMmIiIiIn2j9cA3JSUFDRo0kMFtYVasWIHJkyfj448/xokTJ+S53bp1Q1RUVP45DRs2RN26dR/aIiIisHbtWtSoUUNuRERERFR+aT0hQ6QfPCoFYebMmRg7dixGj1Za4M2ZMwcbNmzAvHnzMGXKFHns1KlTRb7/0KFDWL58OVauXInk5GRkZWXB1tYWH330UaHnZ2RkyC1PYmKifBTvE1tpy/uMsvgsKj28j4aB99Ew8D7qP95Dw5BVivexuNdUaTQaDXSESHVYvXo1+vXrJ59nZmbC0tISq1atyj8mjBw5EvHx8XI290ksWLAA586dw4wZM4o8Z9q0afjkk08eOr506VI5FiIiIiLSvepdQ4cORUJCgpzg1NkZ30eJiYlBTk4O3NzcChwXz8VitdLw/vvvy9SK+2d8K1WqhK5duz7yF7Ikf2LZtm0bunTpAhMTk1L/PCodvI+GgffRMPA+6j/eQ8OQVYr3Me9f6B9HpwPfkjZq1ONb6ZmZmcntQeIGleVvtrL+PCodvI+GgffRMPA+6j/eQ8NgUgr3sbjX0/ritkdxdnaGkZERIiMjCxwXz93d3bU2LiIiIiLSPzod+JqamsLf3x87duzIP5abmyuft2jRQqtjIyIiIiL9ovVUB1FpITg4OP95SEiIrNLg6OgILy8vmW8rFrM1btwYTZs2xQ8//CBLoOVVeSAiIiIi0ovA99ixY+jQoUP+87yFZSLYFVUYBg8ejOjoaFl+7Pbt27Jm7+bNmx9a8FbSRF1hsYnFdURERESk/7Qe+LZv3x6Pq6g2YcIEuZWlgIAAuYlVgnZ2dmX62URERERUznJ8iYiIiIhKCgNfIiIiIioXGPgSERERUbmg9RxfXZeXf1zcjiAl0dVEtN0Tn8ci3fqL99Ew8D4aBt5H/cd7aBiySvE+5sVpj1s3xsD3MZKSkuSjaFtMRERERLodtz2qKIFK87jQuJwTDTMiIiJgY2MDlUpV6p8nfmIRQXZ4eDhsbW1L/fOodPA+GgbeR8PA+6j/eA8NQ2Ip3kcRzoqg18PDA2p10Zm8nPF9DPGL5+npWeafK/6H4G9u/cf7aBh4Hw0D76P+4z00DLaldB+LU36Wi9uIiIiIqFxg4EtERERE5QIDXx1jZmaGjz/+WD6S/uJ9NAy8j4aB91H/8R4aBjMduI9c3EZERERE5QJnfImIiIioXGDgS0RERETlAgNfIiIiIioXGPgSERERUbnAwFeHzJ49G1WqVIG5uTmaNWuGI0eOaHtI9AhffvklmjRpIrv6ubq6ol+/frh06VKBc9LT0xEQEAAnJydYW1vj+eefR2RkpNbGTI/31VdfyS6Nb775Zv4x3kf9cPPmTQwfPlzeJwsLC9SrVw/Hjh3Lf12s5f7oo49QoUIF+Xrnzp1x5coVrY6ZCsrJycHUqVNRtWpVeY+qV6+O6dOny3uXh/dR9wQGBqJPnz6ya5r483PNmjUFXi/OPYuLi8OwYcNkYwt7e3uMGTMGycnJJT5WBr46YsWKFZg8ebIs83HixAk0aNAA3bp1Q1RUlLaHRkXYs2ePDIYOHTqEbdu2ISsrC127dkVKSkr+OW+99RbWr1+PlStXyvNF++sBAwZoddxUtKNHj+K3335D/fr1CxznfdR9d+7cQatWrWBiYoJNmzYhKCgI3333HRwcHPLP+eabb/DTTz9hzpw5OHz4MKysrOSfs+IHG9INX3/9NX799VfMmjULFy5ckM/Fffv555/zz+F91D0pKSkybhETeIUpzj0TQe/58+fl36f//fefDKZfffXVkh+sKGdG2te0aVNNQEBA/vOcnByNh4eH5ssvv9TquKj4oqKixJSEZs+ePfJ5fHy8xsTERLNy5cr8cy5cuCDPOXjwoBZHSoVJSkrS+Pj4aLZt26Zp166dZtKkSfI476N+eO+99zStW7cu8vXc3FyNu7u75ttvv80/Ju6tmZmZZtmyZWU0SnqcXr16aV5++eUCxwYMGKAZNmyY3Od91H0ANKtXr85/Xpx7FhQUJN939OjR/HM2bdqkUalUmps3b5bo+DjjqwMyMzNx/PhxOfWfR61Wy+cHDx7U6tio+BISEuSjo6OjfBT3VMwC339fa9WqBS8vL95XHSRm73v16lXgfgm8j/ph3bp1aNy4MV544QWZeuTn54c//vgj//WQkBDcvn27wH20s7OTaWW8j7qjZcuW2LFjBy5fviyfnz59Gvv27UOPHj3kc95H/RNSjHsmHkV6g/g9nEecL2IhMUNckoxL9Gr0VGJiYmRek5ubW4Hj4vnFixe1Ni4qvtzcXJkTKv6ptW7duvKY+I1uamoqfzM/eF/Fa6Q7li9fLlOMRKrDg3gf9cO1a9fkP5GLlLEPPvhA3ss33nhD3ruRI0fm36vC/pzlfdQdU6ZMQWJiovzh0sjISP7d+Pnnn8t/Bhd4H/XP7WLcM/EofmC9n7GxsZxIKun7ysCXqIRmC8+dOydnJki/hIeHY9KkSTKvTCwsJf394VPMFn3xxRfyuZjxFb8nRU6hCHxJP/z9999YsmQJli5dijp16uDUqVNyUkEsmuJ9pJLAVAcd4OzsLH+yfXCVuHju7u6utXFR8UyYMEEm4u/atQuenp75x8W9E2ks8fHxBc7nfdUtIpVBLCJt1KiRnGEQm1jAJhZiiH0xK8H7qPvEanFfX98Cx2rXro2wsDC5n3ev+OesbnvnnXfkrO+LL74oq3K89NJLcnGpqKIj8D7qH/di3DPx+OBi/uzsbFnpoaTvKwNfHSD+Kc7f31/mNd0/eyGet2jRQqtjo6KJHH4R9K5evRo7d+6U5XfuJ+6pWGF+/30V5c7EX8S8r7qjU6dOOHv2rJxZytvEzKH4p9W8fd5H3SfSjB4sJyjyRCtXriz3xe9P8Rfo/fdR/JO6yB/kfdQdqampMq/zfmJiSPydKPA+6p+qxbhn4lFMLoiJiDzi71Vx30UucIkq0aVy9NSWL18uVzguWLBArm589dVXNfb29prbt29re2hUhNdff11jZ2en2b17t+bWrVv5W2pqav4548aN03h5eWl27typOXbsmKZFixZyI912f1UHgfdR9x05ckRjbGys+fzzzzVXrlzRLFmyRGNpaalZvHhx/jlfffWV/HN17dq1mjNnzmj69u2rqVq1qiYtLU2rY6d7Ro4cqalYsaLmv//+04SEhGj+/fdfjbOzs+bdd9/NP4f3UTer4pw8eVJuIrScOXOm3A8NDS32PevevbvGz89Pc/jwYc2+fftklZ0hQ4aU+FgZ+OqQn3/+Wf7lampqKsubHTp0SNtDokcQv7kL2+bPn59/jvhNPX78eI2Dg4P8S7h///4yOCb9Cnx5H/XD+vXrNXXr1pWTCLVq1dL8/vvvBV4XZZWmTp2qcXNzk+d06tRJc+nSJa2Nlx6WmJgof++JvwvNzc011apV03z44YeajIyM/HN4H3XPrl27Cv37UPwgU9x7FhsbKwNda2trja2trWb06NEyoC5pKvGfkp1DJiIiIiLSPczxJSIiIqJygYEvEREREZULDHyJiIiIqFxg4EtERERE5QIDXyIiIiIqFxj4EhEREVG5wMCXiIiIiMoFBr5ERFQsKpUKa9as0fYwiIieGgNfIiI9MGrUKBl4Prh1795d20MjItIbxtoeABERFY8IcufPn1/gmJmZmdbGQ0SkbzjjS0SkJ0SQ6+7uXmBzcHCQr4nZ319//RU9evSAhYUFqlWrhlWrVhV4/9mzZ9GxY0f5upOTE1599VUkJycXOGfevHmoU6eO/KwKFSpgwoQJBV6PiYlB//79YWlpCR8fH6xbt64MvjkRUclg4EtEZCCmTp2K559/HqdPn8awYcPw4osv4sKFC/K1lJQUdOvWTQbKR48excqVK7F9+/YCga0InAMCAmRALIJkEdR6e3sX+IxPPvkEgwYNwpkzZ9CzZ0/5OXFxcWX+XYmInoZKo9FonuqdRERUpjm+ixcvhrm5eYHjH3zwgdzEjO+4ceNk8JqnefPmaNSoEX755Rf88ccfeO+99xAeHg4rKyv5+saNG9GnTx9ERETAzc0NFStWxOjRo/HZZ58VOgbxGf/73/8wffr0/GDa2toamzZtYq4xEekF5vgSEemJDh06FAhsBUdHx/z9Fi1aFHhNPD916pTcFzO/DRo0yA96hVatWiE3NxeXLl2SQa0IgDt16vTIMdSvXz9/X1zL1tYWUVFRz/zdiIjKAgNfIiI9IQLNB1MPSorI+y0OExOTAs9FwCyCZyIifcAcXyIiA3Ho0KGHnteuXVvui0eR+yvSE/Ls378farUaNWvWhI2NDapUqYIdO3aU+biJiMoKZ3yJiPRERkYGbt++XeCYsbExnJ2d5b5YsNa4cWO0bt0aS5YswZEjRzB37lz5mliE9vHHH2PkyJGYNm0aoqOjMXHiRLz00ksyv1cQx0WesKurq6wOkZSUJINjcR4RkSFg4EtEpCc2b94sS4zdT8zWXrx4Mb/iwvLlyzF+/Hh53rJly+Dr6ytfE+XHtmzZgkmTJqFJkybyuagAMXPmzPxriaA4PT0d33//Pd5++20ZUA8cOLCMvyURUelhVQciIgMgcm1Xr16Nfv36aXsoREQ6izm+RERERFQuMPAlIiIionKBOb5ERAaAWWtERI/HGV8iIiIiKhcY+BIRERFRucDAl4iIiIjKBQa+RERERFQuMPAlIiIionKBgS8RERERlQsMfImIiIioXGDgS0RERETlAgNfIiIiIkJ58P/v6OeCtHp7hwAAAABJRU5ErkJggg==", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.rcParams[\"figure.figsize\"] = (8, 5)\n", "fig,axs = plt.subplots(1,1)\n", "# Plot training/validation loss\n", "axs.semilogy(history.history['loss'],label='training')\n", "axs.semilogy(history.history['val_loss'],label='validation')\n", "axs.set_xlabel('Epoch')\n", "axs.set_ylabel('Error')\n", "axs.legend()\n", "axs.grid()\n", "plt.savefig('ML_illustration.pdf')" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "We see that both the training and validation errors are decreasing indicating that the ANN could still be trained further.\n", "We can now import the new datasets and scale them." ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "data_interpolation = np.loadtxt('Data_Example_3_interpolation.txt',delimiter=',')\n", "data_extrapolation = np.loadtxt('Data_Example_3_Extrapolation.txt',delimiter=',')\n", "\n", "scaled_interpolation_data = scaler.transform(data_interpolation)\n", "scaled_extrapolation_data = scaler.transform(data_extrapolation)" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "Then we split the data into input and output." ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "X_interpolation_scaled = scaled_interpolation_data[:,0:3]\n", "Y_interpolation_scaled = scaled_interpolation_data[:,3:]\n", "\n", "X_extrapolation_scaled = scaled_extrapolation_data[:,0:3]\n", "Y_extrapolation_scaled = scaled_extrapolation_data[:,3:]" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "Based on the input data from the interpolation and extrapolation sets we can use the trained ANN to make predictions.\n", "The predicted data are plotted versus the expected ones." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\u001b[1m12/12\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step \n", "\u001b[1m5/5\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 8ms/step \n" ] } ], "source": [ "interpolation_prediction = model.predict(X_interpolation_scaled)\n", "extrapolation_prediction = model.predict(X_extrapolation_scaled)" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.rcParams[\"figure.figsize\"] = (15, 10)\n", "fig,axs = plt.subplots(1,2)\n", "axs[0].scatter(Y_interpolation_scaled,interpolation_prediction)\n", "axs[0].plot([0,1],[0,1])\n", "axs[0].set_xlabel('Measured')\n", "axs[0].set_ylabel('Predicted')\n", "axs[0].grid()\n", "# Plot predictions from ANN\n", "axs[1].scatter(Y_extrapolation_scaled,extrapolation_prediction)\n", "axs[1].plot([0,1],[0,1])\n", "axs[1].set_xlabel('Measured')\n", "axs[1].set_ylabel('Predicted')\n", "axs[1].legend()\n", "axs[1].grid()\n", "fig.tight_layout()" ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ "We see from the testing datasets that the interpolated data are rather close to the measured values while the extrapolated data are far off.\n", "To improve the ANN prediction we could:\n", "1. increase the number of epochs for training,\n", "2. increase the size of the network\n", "3. increase the number of data for the training." ] }, { "attachments": {}, "cell_type": "markdown", "metadata": {}, "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": "3.10.0" }, "orig_nbformat": 4 }, "nbformat": 4, "nbformat_minor": 2 }