sine_model/1_variable_regression_model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0kEuMOjn8juu"
   },
   "source": [
    "# Simple Regression with Neural Networks\n",
    "\n",
    "**Install all essential libraries**\n",
    "* tensorflow is our general purpose ml library with all the utlities we need\n",
    "* numpy has all our various data types like arrays, matrices etc.\n",
    "* matplotlib for plotting and visualising the numpy array of values\n",
    "* math is just for math functions and mathematical constants"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "executionInfo": {
     "elapsed": 4,
     "status": "ok",
     "timestamp": 1691561699294,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "9Tdh7cAA7PRx"
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras import layers\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import math"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "63RP2rjr86eO"
   },
   "source": [
    "**Library diagnostic information**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 1040,
     "status": "ok",
     "timestamp": 1691561766327,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "NJ3awkYq71bd",
    "outputId": "adac2355-093a-4c60-da6c-2fa131fd602f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Numpy 1.26.2\n",
      "Tensorflow 2.15.0\n"
     ]
    }
   ],
   "source": [
    "print('Numpy ' + np.__version__)\n",
    "print('Tensorflow ' + tf.__version__)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bWNBci118WRp"
   },
   "source": [
    "**Model Training Parameters**\n",
    "* training - to train the model.\n",
    "* validation - to verify and tune the model.\n",
    "* test - unseen data to see the performance of the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "executionInfo": {
     "elapsed": 5,
     "status": "ok",
     "timestamp": 1691562382865,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "n88G7CoA8Dpd"
   },
   "outputs": [],
   "source": [
    "nsamples = 1000\n",
    "val_ratio = 0.2     # 20% of data is for validation\n",
    "test_ratio = 0.2    # 20% of data is for testing, overall 600 samples for training\n",
    "tflite_model_name = 'sine_model'\n",
    "c_model_name = 'sine_model'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NgbWSF-X8_LM"
   },
   "source": [
    "**generate some sample data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 447
    },
    "executionInfo": {
     "elapsed": 1551,
     "status": "ok",
     "timestamp": 1691562098328,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "G21q9Xk79Dpm",
    "outputId": "f8a3f770-2f67-4ad2-b12f-640a07aaccd6"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f561e709f10>]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhYAAAGdCAYAAABO2DpVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAACtf0lEQVR4nO19d5wdVdn/M/duS9tNb5CEAIFAQiAkBEIvgYCAgIiKiAi8IBiU9hONSFGE8IoiqAjIK6DSUZpUQw0tlSQkBJJAEtI72U3d7O6d3x+7994zc095TpuZOzlfP5i7M6c8c+pznnY83/d9cHBwcHBwcHAwgEzcBDg4ODg4ODikB46xcHBwcHBwcDAGx1g4ODg4ODg4GINjLBwcHBwcHByMwTEWDg4ODg4ODsbgGAsHBwcHBwcHY3CMhYODg4ODg4MxOMbCwcHBwcHBwRgqoq4wl8vBypUroVOnTuB5XtTVOzg4ODg4OCjA933YvHkz9O3bFzIZtlwicsZi5cqV0K9fv6irdXBwcHBwcDCAZcuWwe677858Hzlj0alTJwBoJay2tjbq6h0cHBwcHBwU0NDQAP369Svs4yxEzljk1R+1tbWOsXBwcHBwcCgziMwYnPGmg4ODg4ODgzE4xsLBwcHBwcHBGBxj4eDg4ODg4GAMjrFwcHBwcHBwMAbHWDg4ODg4ODgYg2MsHBwcHBwcHIzBMRYODg4ODg4OxuAYCwcHBwcHBwdjcIyFg4ODg4ODgzE4xsLBwcHBwcHBGBxj4eDg4ODg4GAMjrFwcHBwcHBwMAbHWDg4ODDx7sJ18MxHy+Mmw8HBoYzgGIsIsW5zI0xfsjFuMrTRkvPhz28uTMW3OPBx/t+mwjVPzYbP126Jm5TEYkdTCzz8/mL4csNW63UtXr8Vnpi6FJpbcoVnvu/DjqYW63U7OGDhGIsIccitr8M37/sQPvxiQ9ykaOGp6cvgd/9dAN+870PrdX25YSv8+c2F0LCjyXpdDmys3bzDWFm+78O7C9fB2gZzZcaJP725EG7+zzw49ndvW6/ruN+9DT9/Zg48MvnLwrNrn5oNg294Fb5Y55i/csPGrTvhjHveh39+uCRuUozCMRYx4IMv1qPT+r5vkRI1fIE4va7dvAO2NjZr13XqH9+D3/13Adz0/CfaZTkkAxPnrYHz/zYVDpvwRtykGMHkRa2Suyin6oylmwq/n5m5AgAAHnxvcXQEaGLBms0w4eVPYdO2nXGTEiv+9OZCmL1sE9yQsvXNMRYCkCJHU/CQ6bbvbIExd74D1z87xzgNOvAEH7B+SyOMuvUNOPBX/+Wme2raMjj0ttfh01UNzDRb2piTqYud2oWHBWs2w80vfGJUsiDCxq07Ycyd78A9b30ule/dha2MdS55PLMSsPPZJDKUSkXz0haem7kCvnX/h7BucyM6z0l/mAT3T1oE1z831yJl8WDuinqYs7welXb7znSqsBxjwcF973wB+934KsxetimW+l+Zuwq+WLcVHp2yFHY0tcCrc1cVVALNLTl4Z8G6WFQEnmAFm9V2mmoW7BzX/ftjWNPQCNc+NVtYZxIlN0nCSX+YBA9/sARuiHChvn/SF/D52i1wx2vzpfL5UNqXTS25XaKP1zTsgD++sVCbAczExUVQcNWTs2Dq4o1w+yufSefFbsDlgh1NLXDan96D0//8XmqZBgxSyVi88ekaWLBms3Y5t7/yGTS1+HDD84YXa+SiQCb7zUvz4LJHPoJL/j4dAADun7QILnhwKnz3gclmaYsAS9ZvhRlfFiUQLWk5uiYAi9bZMSD0KOfylpZiv70+b01BuiSLDVsaYdjN/4WfPDFLlTzrePHjlfDq3FXUdzJ7/A8emgZ3TlwAl/xjhhY9NMaC1kdRon57tGqNppacEXWrSZDMhOp8SANSx1jMXrYJLv77dDjpD5PiJoUJ7PTPZord8+S0ZQAAMKVNJfDvGa0ugHNXsNUItqC7fB37u7fh7HvtG37uiujWsSqyurKEPP5//jEd/ufv05TK+deM5bC9qQX+M3ulKdKMon5bE1zx2Ey47JGPtL0v8mo/XSlolrJyxy3EUDkf0KRXAABrG3YIJVjH/PYtGHLTa7A5oYbdrG8LpEnpmSp1jMVnq6PfaGWBXQAqiIU7rFZoiXNECuhP6VwpC3TrWB1ZXZmQoj9vxJg2bGsqnjybKDZXcUgKolaFLFm/Fa56YiZ3fTWlynp+1goYddsb8Itn+ZLilfWt6qTZy9KlTkkDUsdYxC0ONAnyRBies7m0sroMiOw6AFpVKhu37npW5jmC6ezeITqJRQXNglABcZ+0RSDXlKTMOtp84DXjtp3NMH+1unr4woenwXOzVsKZ97zPTGNKo5m32Xl86lIzBTpEjtQxFuUALPOT5ay4OfPOKmiYZt4wGwvmNHTuXyfDwbdM1FpAyxGbdxRP1J3bR8dYqJ6ay40nJj+TSnsMjBHdK4RNyOl/eg/G3jUJ3vxsjVJ9i9e32u7saGIvPKYOOyaK2byjCb77wGR4dMqX4sQGEeiCMhvnJpE+xiLhpx8ZZLMcxqLcVmcJ3P36Qjj6t2/B+i149zUAgKltkUCfnr7MBlmJBTkWTEkRMMiaklgYnLS+78Orc1dpR8HctrMZTr5rEkx45dMQY2Fu3p145zuwqn67Ul7Ztv+izaj3+Vn27FhUm+bZmcvhj28sLPxtwpj7gXcXwwdfbIDrBeoULH79n3lw/O/fTpyxaFIhzVisWLECvve970G3bt2gXbt2cMABB8D06dNt0KYEG8tqXHs4b5OI05PCtOj6s9Wb4aWPixb3f3h9ASzduA3ue/sLpfJ2NSeTuD7XGGNhcDz9d94auOyRj+CYO97WKudfM5bDZ6s3w/3vLAqqQiiNrUr+wrVbYMLL8i6aAMlyN81D9bBz9ZOz4c6JCwoGrSYOTVt2mGUAHnx/MSxatxX+7e7NQUGKsfjqq6/giCOOgMrKSnjllVdg3rx58Pvf/x66dOlii75UArsm8BbuOCUWNpa0cY99VBLqXBQHg4Uo28b3fVi0bkvAziFOREmFsirEMB0kZnz5lVK+t+avDajQmghX2oDEQpkyOhqb1bxMaE0fN6+hO+02bW/17uDN341bd8JLH6+Cnc3x6IKTMs+TjgqZxP/7v/8L/fr1g4ceeqjwbODAgcaJ0gHGyE++TMPlIdNVZNh8X5LHt6q4+NNVDTB6r27a9T/8wRI4ZI+ucOqwPtplifC39xbDb176FL5zSD+4/exh1uujIa7AUlGqXVTg+z4s/2o77N6lHXdd+GRlPVz4UKur7JLbTy15H1CbU9o6jg09iRILU55qPGnsN+/9ABat3wrjjtvLSF0OdiAlsXjhhRdg5MiRcM4550DPnj1h+PDh8MADD3DzNDY2QkNDQ+C/coPpdduExCJNqhAbGPfYR5HUc+fEBQAA8MQ0ObuOlZu2wycr7brJzV62CT5fa8+QNexuqgobhwEAgN++Nh+O+u1b8BeBSm3hGv7dNyR94Vn3n9krY4klY0oNZRKmGFze0raozYj05TmrjdRlGrL2QphYFzy8NX8tPCW59kQBKcZi0aJFcO+998KgQYPgtddeg8svvxx+8pOfwN///ndmngkTJkBdXV3hv379+mkTvauAqwqJk7FIk4WsJlTX0sNvfxNO/eN7sGzjNm0a1lGMXNdtboQz7nkfxtxpL1Acx7Y4Ebi3jaEQhRynMUjkJkm+JcX0Hy39Cn78+ExuhMXG5hY47U/vwvhn2Pf9vDV/LRz/+7fho6V4NQ5VFRLzvDR1AJNVZeY35xWbtsMP/zkdpi7eWBaHHxO48KFpcN2/P7Z6gFCBFGORy+Xg4IMPhttuuw2GDx8Ol156KVxyySVw3333MfOMHz8e6uvrC/8tW2aXuyqH8YQ9ofFSpdErJNwsUYn4P1vdAE9OWxqLSuEzSdfY5pZcIGzwSx+vgpPverfwd/4Tln+lz7CQoA1Zc14h8ULqM4gh8jnilt935q+DuSsauDEZLnxoGixatxXO/78paDKoIb0NNKTO/Rb5NWnphm3SHl0kZKdhPv01T86C1z5ZA9+6PxlRfTGfYYoZXCtxAVwUkGIs+vTpA/vvv3/g2X777QdLl7InTXV1NdTW1gb+swmdyeX7ftnEd4/TxsLWaYC3oNj83JPvehd+9u858NIc+l0QPOiKMkls29ksvEb6lLvfhf1ufLUQxvh3/w2exvP0RDE8TKlCTEKFIhlbEVG7zl+9OcDUyfTDVolNffayTfDqXLPqgF//Zx7sd+OryheD+QCwdvMOOPqOt2Dkb17H5aE0EEbNSzsELP9qO/EeVX3sMLl+JAlSjMURRxwB8+cHF7IFCxbAgAEDjBKlA51N79qnZ8PQm15LzI17vCGXZImFKcqi/sJPVsrrynW7gVwgh938Xzjo1xO5N9YubDslT1/SKjbXGe9/mLgAxj8zR1lSk5bImyJDyCmLi95KoqYae9ck+B4hedD5tMemLIXjfvc2LN1QKn364IsNcNkjM2DRuqLURLcZH3x/MQAA/H6i3G21eeRyPsxdob92hte2xuYWWLlJLd5H5Egerx0LpBiLq6++GiZPngy33XYbfP755/DYY4/BX//6Vxg3bpwt+iLFMx+tAIDW66BtwsRCmjZ3UwCaKkSvzrfmr4Wrn5xl9Wp53V4g8+fdaxdIqEfCm2K+zTDtdfcbC+HxqUthgcB4kYUsx2uJh/DQjXstJtuQxmRd9kjREJg8YbLoXkIwAjreG794dg4sXr8Vfsm5XXl1ffH6dVMMGk1igCk75wNs36nvBhpe2065+104/PY3pZiWuJnVXR1SK8MhhxwCzz77LDz++OMwdOhQuOWWW+Cuu+6C8847zxZ9saCqwm5AUh29Wn7CxHrVeAyzVuVrL3xoGjw7cwX8oc1zQxfzVjbAMXe8ZfQWTlX+ML/BmRAa0C7WwoB2w2Y5grQVEU0rsr8wXSfivTD937CdzRjbUEeJ1pbG5hb4v3cXlRgM+uBr3/4KUNoHi9qihopUlUljJhIsVLYO6aXhtNNOgzlz5sCOHTvg008/hUsuucQGXcowYQxTqXgSMw3a6Skvfo6Cr/B9P7YYCQDm9I8rvmKLUWW+76f/mg1fbtgGP358JlGADmUMmtr+bW7JweRFG7gGdSUSC/PkoOsWYWdzDqYt2QjNIUbGlrtpsXz8+2bBJTyy7Sv6Nkx5jZxgUCRTZKodaYHpyJLvf2cR/OalT0s8jnI5gB1EwC/VtQNlY6FUsj5Ug/btakjGDpowVFbEu9DxEJX/uu/78O37J8O3/zq5ZIGI6mBgzn2N/e6ap2YXfou+i7bg2TS++vNbn8N3/joZLvlHacj8gsqDoT+KYvmT3chufH4unHPfh/D0jGBYZNsnTREDRM6p5raIm6yxd/zv3oYlbbEUMGTzLhLEghedk/w2U81IV4UUS2e5xeZ8P8AEm96DA9IiQdm2htRvXvoU/j0jeWG943Y1DiN1jIWJRaoywTJeVjTOOcvr4W/vLUZx+xu2NMK9b38Baxt2MNOs37ITpi7ZCFMXb4SvtgVFsVGJHE2tSzx7lGdnrij83rh1J/z9gyVMz4z2VVlDFJFg0/bI5C8BAOC9z9eX5mrLpspnBuI0RNSfskHElED5FlEbkZt//kTKYhgbm3Pwi2fZMSlKyDHQto2cG0VtHDREEgvWdPL9oHRFJP3h1SEC2T/5XzZDr5O49unZ7JdOoAEAkiG9dxWYtLH423uLYePWRvjp2MHSeWljlLWOnP7n9wAAoFNNBXxrJD8I2Y8e/QimLN4I/5m9El6+8ighHcniheWBtUfJb3xvfLYW/nHRqJL37atKp0tcmqJ8taqqELJJFq/fClMWbYTzDusP1RV05ok2BuJUk8lAJFkh32PGymaJC65EhqEY5DdrWv6ARMTQRKUF38Ns2jk/aGMhyVcIEWAmymPo7bJwjAUFVQYlFre8OA8AAL5x8O6FZzrzv0JA23yER8GUxa3Xi89bpRaK2JbYLVxqXJH8Ji1YBwCtmwx5ImxHkVhoe4VoFmBCYnHFY602I1sam+EnJwyiptHBMx8th1c4MRdsM66iNpKxsQCQM3YNXrmOzhZAY9tmTeN5SAGmqXmpakfgQzDAFuvuEBMhqOm3zCbrCIRRk6aVQUquzF8RJgyYeJd/qYK8xrccbCxM4flZK9AblK05puJBM3vZJtjvxlfh/neKrsc0VYju5kvLLVNk2CuAl3fBms2wYM1mZr2yN4PS6vqUwqxe89RsmDhvjVTZJiGysSC/Iz9WeO0ow1iQdate0pWXWNDGsY31oIXCXAWvjqd/R873A8abNHpbcj5c9++P0bRs3MoPGMeiJ65VMq0Br2SROsbCBMLGmyYGC3kKwHLWtPkrCkoUBQcswxhd+cQs+GjpJsWazHyMSsyP8c/MgZ3NOZjwymeFZzZsLJTdTf28uymuM3Y0tcBJf5gEJ/1hEuxoaqG2SXhoiWijzYuLH55W+P3fT1bDsXe8JSbOspGHmLEofkdzi7hDmhBpaHWrxp7Z2cbI0PKTqhDdqMN5UCUWiLJ9H4DkuWiMhazdxSl30++6KRc1nAhJc5E1hdQxFqr9ROoVTapC8jAVdyIJ1yWLKAjP+frt4lMHrVxjqhAFXS+tmdtVFjWH+YUNQ+LKTdthpsQFUxgUbSzCz+kUkSLqhh1NdFGygbFFXoh26T9nBIJFxQXRZ5FNgZmnchKL4m9dmwMaY2HKxZRklqg2Fogycr4f+F4qYyHBlAEArGmg34FB5X3iXxqlYWqNS9q3p4+xUGzgncRiETbeNKG7IycZnsbSUVdh+ErJ7Ttb4K7XF8A8hXDW5QKVkyKtj0iJRX68YIo+/PY34ay/fFBQQ5DASsM2hC51YrmbsuipJMb0Tk5chCBtgveUBCzjTx7s21jgpXz5EzXv2/ObMG9TzzOengGJRR50poeUhKqDXP9E6hXWZ4S/j/a9puJAJPVKg4SSFTlSx1iogpxYYXdTE6oQUxIL4aSXpPWu1xfAXa8vhK/98V1x4jZE5m5qaJKq6LZpzCRpvKlyC+THyDto6GGlZ1DTqqjYm1p8hvGbHGhl1FTKLyn241jw3+dEaoAQdnLiSuRBcwfW3QxFEg+ddiSZTZqNWdArhGFjkRNLf8LB0VRBoyBhh/ZdGqljLDBGRjSQE8uUQRRZv6rhVhimLn7KY/byTdJ5xOLXZLHtNNGuCGQzL2yTNJDjAnvqJ0F32cTlnbYkrEqh21hgimtqyTHE6mHa+KXR3qpILEyCxhAKJRbEb1M2FkVVFSGxoAwZmWWBtoaYYr5F45ls1/c/30BNEx4vdBuL4DMp+hlpTTTBlxu2ojzqZLArSy9Sx1iQkOlYUm8aXoZMqEJyAVWIuvGm6OIn2cGM0fvyisQwb7r3YehCiakj+uiMe94veW1qzdAtB2NzEw7NvrM5x6hXbpzT+l5JYmH5rCkMq00eABBMKMYAkWZcS5dY4EeASOKh044kY6FuZBqUhGEYC1WY3rSPueNtGHvXJKjfpndh4S7MSwSQbsZCIi05scL5VDe4oO7WjC7U9BKsK0mJKnS+Dpkq0leynbe1qT10FzOTIv+ijQX9OQ/NOfodMCVlKdBVU6lgYxGzKoQ2T/nupniJBflt+qoQisSC+O15AO8tXA/fuv9D+GKd3I21O1vIUNzisUHD+i2NsG5z0RaItrboqEJ8zl8AQQZSZkyRc2E1JRqx6iV9uzJSx1gEA9KoqUJMsZ1BfaOZwWl6EWadILDtiFks80lacj785e3PmemwhoiyUHFNE3oSGKZNlkaaqL31OcZGIGeEIaSrQlqXlOVfxe8NkodYFSInscCA1p26TDwm//f+NgWmLt4IPyKuesdgZ3OxbB37m9c/LcYroTFCPKZMjhnAp5UpK0zDPW99DoOuf4War2FHE9z3zheBsZ4WN1hdpI+xIH7LrBGNAYmF+cFBMr3YCUQ1UDLAWAROUYxGIudHOIVqNMHHpy6F3746n/k+PClN9YLKZkG1hwjcT2CeWSmWjQfG5Mb3g2U2teSolfCKoqoTKGVUV2Th+Vkr4Mj/fUtMGKJeExAabxJzUzbOAgu08aG759CGcWBTJJ6v20J302SX7VN/F8pWWHhojJApxo1KI/Fbpq3JssJfecdr7PXqhufmwu2vfAZnUlSluzpSx1iQkFn8ycFl46RsasGSqTOPLY3NsKWRiPxJvGNJLHhx+Uldrox4V9Y4Cls0eT8BDUpeIRE7hsuSWPQ6kNdf7GzJUeeGrFqFVkZlRQb+/CZbKkWD7aaWubq8cLupJltLlVhobqoqRsgqoM0XlS6iGcKaUiv4zD/kQTarzLzPG7Gu30KP0yMiqyXnl9z0mxakjrFQPU3rMhPrNjfCrGWbgmUSv035XcsaaDW35GDoTa/B0JteK0zqQJhhFsMTkFiwacd8Vj6JyHCrRBWCWDGem7kCBt/wKjw65UtmGpUFWeTBoWQPSik1X47q6GAthDyJU1NzzshdC7QyKjJerMF6aHWLIvTLGm/KgGwjG3EsAnMEeeGZ7/vw/56eDX+d9AXxjKDT0BlINo6FkJGVOPzJjMGAxEJz7Mr08MtzVulVRiBprrapYyxU4TN+Y3HIra/Dmfe8D7NDzEUeAVUIliYDaxx5E+NXbdeBk5OHtZDyTgQYn3YabJy4rnpyFgAAXP/sXHa9KkwAopNY16vLlKlsGFxwNw0/Z6QnXjS1+Dh3UwXaVNyhbXuFiGwsyPEh47XAK5U2d3Xns0jyhm3FyYs2wr9mLIfbXi6Gqw9KKDWMLAjQ7woxJLEgmYw87YrDiKVOEkFsh8Xvr03bNT1QEmzPkULGguTa1UrQ6a8PFxV9vIOuV4Q7qwZbbMbGgpRYKNhYEL8x63B+Aki7mhmaN2o2FmLx+VESdgQiSBtvMlQhxYiP7LxNLXR3U9mxRSujtl2lXCERQOySS0os8BFVeUloTJmuNIQ2RlZ8tZ1RPxvbm0qvfQ9KVkrzqCw7NOZVx3iTXLdMbqkknSavTBBKVYzVlDykkLEoQubEReWAEWlL39HrVzHexEAlmBeZg3UK4tlYBNJJbIhxheFVqhdxGtncWLpAy0I9xkcrMBfx+iHrzZ0tOFWIWDRd+qxL+0p5CYTmfNja2MwdhzIePibjLPi+H5hHtHEo0/8084SL/z698BurBqaq5IjfOsabgSvoKUyEDHNVeikef01SHUaqqhCdgHcmkGCBRfoYi6DfOD5fQBXCyffApEUw8jevw+dr6X7iLKZExXgTs2mLROG0iRL0CmHVza5T1MbhvPk/RYuKiTgKNOTr3bazGc76y/twz1ti40L6omFvJquWrOKi29SSk/4WrBon6sVuzvJ6GHLTa3Dt07OZaaRUIRKXZPFKvf7ZOXDk/74FWwhVpJ27Qujg9i+tLwNeIZQsyA2XVIXRDi2qV86HYXIuBow3JdiTpF38lSSkj7EgfssMPmzSW1/+FDZs3Qm/fnGesJywXptGoy6kRHcU8TnGxoJvvIlvY9OGcVjkF/THpiyFmUs3cV3I8rARx0In4mppmtZErP4XGm9SCUSRR60jUEyEC+6977Qyic98tIKZRhggS9ONmIbnZq2EFZu2wzMzi3RRmXCJMk1F3qS7UhO/NTZt8p4R2qFFhnHLZNhMs6iUB95djK7HV5BY5HI+8+bVqJBggQVUiJOUL1Qb3kaHmbp8JzzuS3XscmUw3U05YkdP0Y5FlrEwdSrJV9socb8Hz4NDFTzRqWrZGOPNLTua4Xf/LTJTpi4ho0HlM+zHsRDYyzAOAyaQNXi7qZCxCBhVs0Frj4A9mIa7qUhiIaNq4jGE1PGryNEG3U1xeZ6esYz6HKtClqmrHJE6xiJg4CM1j4Ory6X/mM5OCuyJxtoMmxVGL4Z8kY0FbR6j3E05dMiGKc4nkdVfPzdrJUGD+qKsZLwZ0aT3C/+qfR+Gsbz15U/hX4S/PDuOhaSNBe2ZRUmOKkTl27T9IU/d2nEsZLIjVZm0DDTvLWwfZbP8tUVGJcxVhdDGL7rkIII2FrhSXp6zWrE2c3BeIXFBxsaCSLtxaxP8d94admLAbTxBrxBDgyBUcZiM8IQLBP6iZEJ5hYQGsMiCnAVTrmayUNk4RIEm1TZQSpmKi4Pvt9r7PDuTrQLIY07ounZWSG/phTnChe2dBevgo6VfCUmgfYOMfXNRgiT+NswakCVWWN3mEtEUUANz0mUChy+/7V+yHn7ZPGQDh5bS92FVCO+b+DYW9N8qCAZHxBXGHFMBdY0Z1VU5In0SC+K3lFcI8Tu8EclMNNa4bObcnqoDGVsAGm0NO8SeDaXGmHydNOuJhHrVKNQCZJnpJexCJWNXkcetL39amoba+qHxzKDr01UN8Mana+CE/XoxywqXg6lfhLBdVPjUuLp+B1zw4FQAAFhy+6nS5UupQiTKxfQZaXNg28aI9pm/ffUz2B6KTBt2F8+G8qleQgYQbD/d2015aj5dJj9QrgKTgvkO23x3cuUVKZRYqHqFkNDpsODgL/7VFLg2HVkWgqERia+D1t6tv3FXbePeydlYxCOxUAnp/d7n60ueBTcgeYZB5OZXWracKFgGtCZZuHYLXPz36ejQ67QyfF9etSFyk1xDuXGSlpcF4V0hCidWLMTXpuMhx/T4sKOpBf7y9hfw0PtLgi8pEsvgwUqVQvp6QyJsa0amCM8P7jgKSE11VUzy+d9dWLo+xImoryAQIXWMBQllr5BQPrrLJsIan6EKoW0wTS05OPOe9+H6Z+dw6dQJYpSfQCg1DscIiSfd4SE2rxAL/Az2s6mMg4b6QyUN9lke+Su341Dh2qhSeFcIRWKBMoJGzKOKLMlYiNPzIDYGJFQcwL5Dh8bs+MKNmv6x4TktimLaxGkE5bud8hQq7q0mlyWWVIUG7fDhCRZZpJuxCP/t+zB/9WaqL7VPGajFd6Vls1Vs9N4W+W9PWrAOZi3bBI9OWcpNF4bIK4B2uZqsxKJElC4psciniYuxaJFgqHhQkRLQgu/4MqsPI6m2+JfrQowsg5Jwdf0O+HRVgxQtQYmF+TEiNnBW1IVI1q1vJCp3UNrJ8IIKqkLyjAX7IAHAnjvH//7tUL2ExIIW0jsssTAgVc6XwVJfisYUSWeSN+tyQuoYC96m9/jUZTD2rklw+SMzSvMRvzELAGuisdYokf/2ckpoXozls9RprDABxeBtYqJogiwILyFDlyQHGwwNtkTelcyt5ZSeGDG1aC2APt3dFF87+/2rn+hZy9tY12tr+KZkNveSgLsp7RIxVamqKC34VPdqD4KeKnmSyKJl3E2/3LAtVC9RjqaNBVcTIsGdi9qN9V6XybXttWEq5ooNpM54k0S44f/vvUUAAPD6p2v5+TT6K7ghF/9q5thY7GhqgZte+ES9Ug7Ija2oCsFILHASnHBbTVuyEX706EfB9Pn6Y5JYAACs5ejpsVAZF1Q1BOV9YKz6ABNe+bTkOmZen9DS8NLyPkXFfkQHorgoYYkGb/yKvHmo4EjndJERxHWgYebSr6hG1bKi9Z0MKSmZrGBjQRSuY7xJEomJYyF37ULxd46yBjEPe4Jyg2qVogTnG/d+gKaNRuOujNQxFj7zD/6JWF4a6sEvnp0DWxub4a5vH0QviADPK2TDVrlbMoNUBFFqvFn8LWdjQZaBPzmf98AUZloVI0pTMHlFMQD+NBIw3pQQydz/ziJJiuTAkzQVmB2htMUMRLflhhlZWZWWiJ+lqQtNfRsZMAo7/M/6C31DE9pYhBg0piqE5m4asqkSMXAskG0pKylU9cTCMA6XPjwd+nVtDzd/fQj1faGstp+bG5th5tJNSvRg6dKV0CaZiUmhKoQ8oQff8SaKTMQ0gNZAL49NWQrPz1oZUGOwBHQ8iQVzk0LQQfukSQvWwXf++iF8uWErlctHqUI4jBbP8pt2SiraWCAqtoRetTXaLqR44SuZDif+NqGaYKWh1fflhq2I0gR1WVjZhGJrpTIFonLKb1OfJrKxsLU3+MBjLIq/W2i6ECj9fuzcIddcGmPBO/ioQlTGrGWb4I3P1sLDHyyhvo9RkGoMCXMKSR9jQSK8qGMD5WDEc7jrwsnf+FM/D6XMUWmIrO8/OBUmL9oIVz05i+pKh7tfhM1oqdq6xeVuCgDQuX1VLPUGxwnO3RQDnU39nQXr4KKH+ZFlAezaHqjCNjNj8xRo49p0EoFp7bNVISQYfEVh3WhqycGD7y2G1UhVoqwqyURzi+psEoTyp6kYVfdpF9K7FaljLIIi/OA7LtctOcJZkxwjfMBy/xiSeINz49ad1NOY/ID2mX+ZNN7kUqC5AincLs8lAksP1SuEVrRc9RwbC3HZs0OROFVoMQmZMPwqtInGKE3HjgFmHpn0OJAVrbMkFrQDT5i2PNn//PBL5oWLNOQQY1QVrL7RbdccpT3MwO5MSrIqJHU2FiTC7Y6MtyI9HjIscafiSejztZth+VfboTJbyvcJ5RVEPRnPoxpvSrub8iQWiA/LLwhxGm9aOOfiUok2yvyJEWGYianepPGhqjW9Vp0C+lVcNlWYFVPtSJaiL7GQqddnMhYkVXm7J9ZVAB8v3yRDonDw8lzXW3xfSUVXPDAx3E0F+U3eFZPkzT5KpI6xkN30qPlC7z5Z2QDPz1oBZxy0GzUvJqaFQCIewJg7JwEAwPhTBvMTAp9Z8rzgd8loImiSjuLfaqcwHYlFEqBCPc0AMjjWJMao6mZnSBpXWqyZ/hSxuaLInMGyKOomEZkKkigsRN4WUmUJ2jvcTqzbfHN+6e9SiYUarcGDlVwZl/1zRkn4cQx0pQzU+5QMQGWsivDaJ6thZ3MOTj+wr3EPJpNIHWPBsw3AWjnTBuqVT8yC3rU11LIDl/pIlMmiM4/PKKGVw58QHpxhuqghvREKMKzEQsbWJFaJhcUNQzYdjTHDFIdhSDCqEFOwIrGwUKZYFULUb7juYMwX7cK4CHiFANvGInjY8AvpSbBsLzRJ5PavClMhKlM1f9JCZAO02rv88J+tMZiO2Ls71FQm15IhuZQZQHjA8PTsmAVg4dotRBriJMo4UWGNHHU4T974z3ghyUPbHxhOmRuZkfht1MbC+vXZevlVFrDwlcyPTvkSpizaqEdIVIiBD9TZmNTLVD9lC+s2KrEQvQ+maGRs1LTDRvjgQwv1jaJR+vSv397572ZN7yenLSumpXwQzd1UnZbg7/VbGmH8M3NKbhhWAalK29rYnGi1S+okFrxTHT+SG/13IA0jr8gPv7VMeYkF6urmcB6i/rCNRWECUtqhUyg6Ifd0rDgR4wrpDWBObF8sTz7dlEUb4P/eW0x9L7+A69ElW7ZS4CkkZEJ6N+dy8OZn6+Hg/l2onj7UfhYQGt4MWung58FCdClXqTRQfp1g1cuUWBC/aZeQAQD4ig5cNqc4e53k53th9spA2vBYDtJs9gPGPzMHJs5bA49PXap0My8J3ncmTb6yi0kseHEs5Aokyw5IABhSioB9A1aMjkjDj0LolagtNm3bWRKGFwBgUM+Owbo5jJasxCKfIs4AWSagYltCts+XG0vbXZ0WxnPavirZ7oXog4IRaMd4k48H3l0MFz08nRlEigYpVYjhb6Jt4tz0GvWHJaQYr5ACSaF6CxILdXJQ32KCESlIYiUD/xVpsLMu+T7AwjWcm4K17y5KLtInsWD8BsD3I1PqwEjj0fkKtkuqjupDYjRmM6UhvW94nh46PJ+quSUHD3+wBOraVRbfsQUWUmgR3JdiFZpVr65XCwkulIRJGFmotruNVvd9H9ZvaTRSliikN4kX206fi9fTvQewV9MH3ltkeE1qWYSMXqhelvEm7a4flldIOcCHVqnEJyvFl9/lfB+yJXZppQcG5TgWFAmxKSTQ7IOJ9DEWgYkc6liexAIh3kepSxSYCZPDr9R4k3znC93HHp+6FH7z0qch+nzm3yiJRVuacpZYXPHYRzBqYNfC36bu02jJ+fC71+bDXj07oPPw0+i3cYHX4RT16xfnwUPvL9GuC6A0sBMPKl8njLxJWTNMjVRynEz8dA0iPeedqG1C75kSkoDEwqfmNSNJKC0k/MwEU+f7Pvzk8ZnItKXPaMa7pvqfK1GWLEtGbRY3UsdYkJCRWGAkDSy3JIzBJk/cKhNsa+qSjXDVE8VJxHc3LVWFiJgmGtfPk1jILEBxupvq1vzxivoAY4FFkPEqpeKVuathymKcMafcCchuW5tiKsIQq1/kv+ujpZtgR1ML1FRmhXXKlI4ygiYKfOnj0vtqwvVxbSywhAnKIp/SLiEDYBt1JhEyFNIOQra81Xy/PNrPBqRsLG6++WbwPC/w3+DB4lgLUSKoBw92KlaUxBoKXGkIJS8mPQB7c2bleG5W0RhJ5BUStniW8YVnAcOECTOWGTxg9y0PojyrVFUsTCNh3DNcHdEgILCwILEAAPjD6wvYZSIkkXRaxIll96yT736XXZ9I8hKWLEpIXsNJjdhYCOrWLd9UIbRDnypDIJYq+TBtyUbYvKNJvuwSyXFyIW28OWTIEFi1alXhv/fee88GXcrgLRLciJOIxYXJNEguMDKnFBF4p6aM56ENLWVEsKoSCxFmfvmV1dODDkovjsPlE32PjN4Up4rDl4eppwh7Cl6WjRINqt/32tzVzHfUuWnopCmrZ/+ccGkvLUtQV4nUQUxTUWIRTFxOsexk2pjumUNKrMx9OG0IPT1jOZxz34dw5j3vJzJWhilIq0IqKiqgd+/eNmgxjnC/8lUhxdSsDRjjo81iOHibOnMBQCxuJRse8Tssscj5PmK9pBm/hTlltlRIB8/MXAF7hbxTkgIPPKW9hqUaK5YrURbjtzifHOH51MG+jWanCY+nu15fAP/48EuCCjYdX23dyS6XV6fieEYZUptsNsmyMFItlmSiwBBr0K9nK2S2Hl5a2vprqtvCzMMLbdLmL9YphC7nHPCSBmnGYuHChdC3b1+oqamB0aNHw4QJE6B///7M9I2NjdDYWLQeb2gQW+7qgCd6xsaxwEWTpO8YPtCf82wsmIyMmAzu0ha2sWjV+SEKDdNRImIJ/szlfPA8Tqx+iTpnLt1ELwNfBCO/b/TMjTbeFKqexIycDIp58Z4W8YMdufau1xcG/uaFpR9+y0SoqqALYfNzbOPWndC1QzD+hc32MctXSDKICMkr+xAlVRUaOsWybd/wZVBtLELqYlOIK3ZOEiClCjn00EPh4YcfhldffRXuvfdeWLx4MRx11FGweTPbV3fChAlQV1dX+K9fv37aRGNRGiBLPaR36/Pib7aUQa7M1rI0VCGhbyKLynjBern1SKhJyL93Nufg+N+/Dd9/cKqQ1iTHthehNYqp/AJk8uZEuQiRGsyJAf26ZI1E3XolsWI35HIAj075Eg6+ZSLcHWJWqN4LemRwy5Z5r1MXW6pKrgn5Z8E0LDdUKXqUc8rWg6+JlpLGWEQtScGVFxZZmC3fJKQYi1NOOQXOOeccGDZsGIwdOxZefvll2LRpEzz11FPMPOPHj4f6+vrCf8uWLWOmNQGe2yhXFYJYs1nW48wJzCofaWSFAe+bwjYWmHqoERbDCxbx96xlm2DJhm3w7sL1cAvjemUTDIWutEHfxkKNAtHGoapmZW8auGcqZUcBG14heVz/7FwAKDXkVDXexMCkrYJ0PyIkoUWX1GBam0Gjgn/r1yOlCqHwnUalFBJ7jrS7KZk3fGdUwsw1tNxNO3fuDPvssw98/vnnzDTV1dVQXV2tU40ywpMDG3kTZXGPscMISDg4EgEMZ8JCiY1FMVM24wWMB1ttLOTVLiUSC8Yp/G+hcNUqYHVRIphzBSJEF1zRPhcjwo4CUalQZEw5VEnCunHKMMEoD6oIGQu0Dp6yLmHVszLAMA0mmkemDKEqJF+aoX4zOYVKBRaJWBWp0ArpvWXLFvjiiy+gT58+pugxCjkbi1LxYEka1m/B5tFaJluSYut0sHTjthI6RTVhNjpZKYgMkqjbBdA5EYgkFvKRIgEsMx8GpB4GqzYCnuRAVRSOSWNybotKKjkAMHIEDLpz9LwsFUkiIUEkLaXNkO486EgZkt4vUozF//t//w/eeecdWLJkCXzwwQdw1llnQTabhXPPPdcWfdLgNTi2IzEGTWxbCuI3MYx5J1e28aZ49PCkMF9u2AZvfbY2UA+Gbhol+LSyZZcHVONYkP2+rbH0tkkprxBDm52wjLYvjeJEtHj9VvjVf4oqNNlTORZYiWGSh6qs2gBjByaSWBg/OIRbWKJ8HcFuHiKJhS5EdlhaKgusRCoBkFKFLF++HM4991zYsGED9OjRA4488kiYPHky9OjRwxZ9WijVdyEvIWOeBoNqBVGaIC3sUaCjhy35olBZj09dKlUP3cYiXAUpOjQLW7pC3/eNlo3ddMlxMnVJaYRNjE0LUak4DS2b5Aq0M8I7Xc6+9wPYyHETDUN1E+DlUv3aqPXaQokFcuMhHzPvCmmTZMxatglNX1yQsrGgSuNKH8p74PhU6aMtT7Sk2VSEIcVYPPHEE7boMIbgphccHOhr0zFpGJIJVvrAps4xhjSNsD2Eyim0RMQqKTpMMmeNhbrxJv/9gjWlAZF0movWv7Ll3fDcXOjcrhIODdyNYgdhpkLmoi0Z8PqBZhdlSlpj6jScy/nww3/O4KbhxZsJPA+ofdkSiznL62F1g1pkWFqZtGe81kFfGqnJZJu6YbU0phBlLhKPZJcUmbaLG6m7Np236fE3B7E0ggRKXRJ4Li+xwMwXXoCs1rLJ7+KdYnCiYpl8MuUkHZ5H34BEUNpYEH0kU6oKS/Rj5KVOpmFLFcI13kQcElRhasxjbu5UkVjsaMrBhi2N1MMDTcKWRMg0MS2tCXfT/AWLsocuGZTT8pnqS8jCizp2gWVt9KzwzDjjTXYanSA1ouh/snEUqNdOcyQsKImFOIl16NKgfo2yQh4EtcxyaadEeRK08ukA8eVK5fIj38ozjCbqlYHKzcCYeDw/evQjAAD4xdeCdz7lfP2AcrhxrH8wkWkauo0FUZbp8WXRLTTJF5yljrHgbfIZrCqEaT9BT//pqgbq8+AGTEgOQgyKihQhj9I7LNji0MvbFhFqXRJj1Gf8NoGkqg5Lopgi85mc+yh1nSUaolrDZK44lypX8Z0uRGW/u3C9tbrYbVX64p0F6wJ/txi2SWLXrA8Z5g1vYyGHogotVIZkQUs3bIOONRUl0WFb6wiv7clF+lQh5O9Qy/NUIZjNkiUyvfjv06m5WTYWN/9nHuxoaiHS2RsisvpDqjFh+G9eI1uCdjWa+cNMKbbPlGxaErJikHSsqt8O9dvlb2TUqZOG8CayfjPO8JNvY0H+LhVp6yDSvkRuPDSawqHSfRMSiwRKM22F9G5RMNQIS4fXNuyAo+94Cw6+ZSI1fUKWBRRSx1gEwVaF8HWuLImFeACynocH9NvzSTdQJilClIT0LqEHuQEiF97WOoh2QJWdjCmBujSKk1vlK1T6FqMbZ+elnb702//KJ2bBgb/6r3Y5uiCZm8Xrt6LtAPjupvonVjbiG/sydxCF0+Z8gAxPxKsKC2uBlLSVSPvhFxvg2qdmw4YtRebUL6STo1N0ayoGn6zi29GE7TcSsqxSkUJVCHvzJ/fgnA+QJa9rJqULjIuOUKJoJmPB/lvPxoKfR3bwUQNkha3NLYijRdAVy+puruu3NMKrxNXb2NJUdOzlxKyZhlhiUfz98pxVRsq1KYAzVR5m+GNVIbTnUxYHGbRczoelG7ahaMPSg6VFvh41Vci5D0wGAKCqHWRRCDRWsukUf9Zvb4JpBCNcEpZbUIeqgX0cSLXEorSti10XXvB5bqq08mSlGqUGkMXfWrebhkbjC7NXBv7GbmzcVCUSC+J3AsWdtrBi0/bCb+xENmtjIS7MZFvHETLYdEjtQrlIiYXpL7YVVZcG9JhEfOW6LY3wfwZC9JsE6qoFZBlkn2/aRkgsFPuLKrEI/X3WX96HHU2c63lF4EiOk4bUMRZkU4eNJEnJXlgnFhYz0UCWJ6sK4aoTLI4PeRsLcYhplq2JCdgLkGWnXHG9ChILSeNhEDxP2mmGBxlaZVRb/JDepfWXpY1FCDqSUFp8FVlQx2LJ3/oNJFNCvq/XNDQWnvXt3K6kLFmqMF47i9Zt5b6XiZVTup8kC6lThYBP/QkAYVUIuytYi1CL5MmGF6kTJbHAqEIicKMoLrZ+W9tEwxTRaEgODEiCNPKIbwHVoyFcRlSQqVLGBAArRjYex8JQOagLz5AqyyRNJSOqEBmJRVviLY3NhWcqhpcsGsLMv8zyLFaFkL/9ZHVkCOljLAiUqruKXVcisQjko/dYUGIhd6osVb2Q76hFoaBnkFgE//bH1nfn3PchrG7YAYcO7Ea8wxSuSZwBxLFJtuR8pUULI40oh01DBTISHjlVCOedRUY5ynHHk4oG0yGIMkC4KPKkKcgwg8XpWMyzbSfpoRf8V48Gs2xq8g5XbKSOseDZSmQIxQ/NvarwjsVYICQWWOkDK8aFLKKUWEz/8isAAKitEUcBpJdTRjNDANGn7GhqgRN+/07gZGS0fpm0is0eR2/J1CnDVHMZC6rI3szXmypH5QChc2CJqu9NjGO58e2X5Nm+s6XkvSxMqNBEa3mYAWYdYpOA1DEWAfitm9kvn5sLA7q1D0osuCd0OloIZgRjlBWUSvBUIbKUFBFFQKkwFbJ+30kwMjJNgQ/5MN/095MWrAsYe8qXTnsqZn4LcRgQ5SURUjYWEoMf625qfJxEKbFA1o1auyypKOJeCwrXwRPPdraUGlTKX0KGe6YDE+rNqJA6xiLc+LOWbYJHp7Te8HnqsD6FdzwRNWZCMgcNM284GUKtEsPowdy2yfN22ZWQ9TxoZny/6qVlAEhmLYJmj6dv7dTJKzXwTuKbcf0UXRuWVqW+rkTmzWKgGtZVC9TqBJIFVSkPq714q0B4jZCTwPmJk1KQSLdXiO8H9GfkSxVjyhaMV0igTPYJEyexiA58HXQQsotOEiaA8WvTfX4AIZ2qNARYRat2S21+zB1vwUPvL7ZTOMhKLGS8QtgFY1Scqohy6LOuPi9Jl4D5aBIynyNau86570P4aOlX0h1HG0FRjqWkHfBSx1iQeOj9JXD9s3MKf+d4Gz3CgAvjUsQSq5baWBR/68SxMD2cqFxzqBKbCzGba0/WxAHgeyXYvmeBuZAYFMnS8n25YRv86j/z1ApEYmdzDpop4ukwTDUx1SjWkIrP1HqvMp4w8XjYee0Aa2BKzYssE1M/r94fPcK+U4kFGalJHuEuFdpYhNbduNVKPKRaFfLmZ2uZ77hxLBgdRg4eWWaAd4GMSF8eN0rc2MjfMqJjQ/SowPjpwfchw1kJdBgLE6qQpIwdWTS1+DB6whtQU5mF9352HDetKeZN9XZLnH1RhOAcXoLPo2GIohqDcl4hrWl5pDXnzGzZos//w8QFyuWVMmjJQqolFmGQG/h/P1nDTKdjY/H8rJUFBoSn7pC9epwFU3M3P5XoNhZ6dZohUW8XaWrJUY20VOEDvx103IAxEQZZBySqi58yJdFj5abtsGHrTlixaTs0NpvrL35vmJmLNEQaeVPwt+h5MI0dukto5FTz1TbcpXcqTczL08LSIXFAY1hEjNWi9cGAWbIrRpLPDqljLP4TCmlNglyMf/3ivIC4FWdjUfzNm3gzln5VSMUqEyMhSQrCzRFUKUVMjCKueGwm/PbV+dFVaFtiIRgzZdItJcgS+qUmASNoQxVSeKaYT6kgS2AyNRHRHZVER6YMDKPXbDD2jEkE60j2DE8VY7Fy03b4cNEGTopgZwRFoPTnwfTFF7w1j+ZxwvMKYRpZsaswDt7E+J9/TIfP124u/M2zVZEtO4xVDTvwiWOE7/M3dxtuwKq366qKpOM4EZEGsW/PX8dNK2e8yXsX1F2bRJSHhpJ+lpBqlaYpIyh48vDaoCUn73FBSy7dhpIB3/zQ30lCqhgL8vpbGrCMKGshbkHYWAAUT108sRjKeNPAyVUWrLH9838XjWB5uj5dzF62ifEmWTNH1O5a7qYaaUTudEkHOU9+/PhMblpTNhaq49mYVAMBVEhvHF8R2digbrYclbCwPA0BTB5YGwtZFMoNPVddB2jtknTJNolUMRbi+xNCm3vgBCg+DWJP6nmjvqAUJFw3SRezqMggIqFJUm1ErSMB32kSfBsLnXIRjGYERrNxLGQy48mYKiTwO7/xmPl2YzZQCuVoeZuV0WQ1zQy25HzpsU+XFEoVEbDLEpXnQ2jPShjTkSrGQgQeI+ozfgfzk6oQscSCVT4AUmLBrMEeWAw2i0nCRSBN1qA3Aj/6/mluEYvsy72tpexrDYkscuEVGwnM5hur8aaEuiwME7ZTdNsVnYJZ66TM4UYssVC730dcrlR5gmdJ5/tSxViIGptnQEn2GiZAFs9vOeuVqkLC6YPhmZlFCWHuRCSS9hR/l6Pxpm0M6tkx8DfPFVUEVlfc/cbCwm/RhmWCwbj79YXiRIYhs6ibklhsayQuoZLIp6OysgG0KgRjYxGjukQWMmsQ9rvM2Fiofx1VFcLbvxKGdDEWhsphXptOvOAFy8pfdsYztssxNuq4gdEJyrhUhdOnBX7h/1oRZiRsBMj690fLi/VLnEZV2/+ZmSvUMmpAShViqI3nrKgveRYVw24TOrZbJhCV9Ezme4p3hZiljdrWsqoQYjyLspbQn7A1Nl2MhWCEoV0+GeUEmAGOyJamCikZeAhL9CRt2iwJy+NTl0VDQMIQ9goxuVDpWO37lATlpC+XMZzTiRWSR/32psBlcVJthUgaZdNjNxucpMUS4exlUKNIs6qQ1jIlaTDMsIj4lKRP6dRE3jz3r5MFrqY0USFdoqArsch6Hvx7xvIAg1HqbkrSFf8oEXLIGhtV/F9Xita7Q8yJFcJNome8aSZNOUImNLKJ7gvHyjDdruakqPKHDNoml/NbQ6bLlqUC420p5KbFoN1uagK0/pGtgxzO1EB3nLNv0paD1DAWn6/bIkzDl1iQv1kSC4Kx4CyAb89fB7e+/Cm37noiqpxKnHlbYK3VSVXdqML3NcNuA9/X3frtpixDNsrCWU69ZdrGoqoiw91IWQu0mNHGeQ6YOjSoLBG0ql+YvRJe4AQRtA0bY1FUZsaTV4FIH54YyVVXAXp57P0raUiNKgTTgTypQSAdYx3CLnrzVjWUPAsPgt9PXACNzS1UumRgbGyZlQDvcgi3iQ0bi0B9Mp1QRh2GuegvD0wb33ja/lL1l7PEwkQeE3nzMD3smB5rQkNmMi0ujyyK5ZY+w0J0GAmr7sNM0spN22HMne/APyd/KVexBaSHscAs5JyOxrh/6kw2GqecD+gVt5FVAIjJKy1hSSBrrUtRq40F+be5b0TpwAVy4QQ2OQpSqhDEcULknVN6wR72RItrY1PCSJx6zNyJ1tb40Zknqu6ztM3e9OeZtkmRtbHwfYBbX/4UPl+7BW54bq5RWlSQHsYCsciUqkKKf2MGBvZuGtrkoS0wtAidgXIiPGpKHYDLdNMiocsI+IIybATICqaRe14ukJFYYBo5q7jCYYz79BhAWSDGROhvkc2ZbSxatwVO+9O78Orc1VbrkWlhTORNzPvScillSK7fQa8Q+TWgscncpX26SA9jgQl5K/i78JzxQmrRC4EmlcjTrCUJMbyTsBg0jESHBRMUJn3DLFWFxBPSWzdt3JCTWIghjCdCOflhELXBtYrkY/OOZuX6TBxqXvx4Fcxd0QCXPTJDuywAM8y0rV6jeZuYlhjxyktaYLzUMBYYcI03ERunSkQ2Wvl55ANpMa/ATtBYCbpWlj/0VSGhqewH3325YWs4i1GwpVz5f4n+StJAEkBmimGCkNFcv00hSu8dDMNlsptt2ZPbGIoym2pxbTdLiOn2ohUX9mJM8qxOjVeIivFmYDMIpKN3GXaBxnZ48U4RVn3Iggwg/21sA6ni77R4hWjl55Txl7e/gDte07iiHaW7L/8+oEHKKwQx6cU2FqG/Dc9xY3ENIqyr3KBiyGx++rRJLAwdwOiRNzUKjBipkVjgIkaGJBbBYybxnA4dVQgP5TRgAOS583L7PhWQn6jFVIDeBkEVyWpREy2MMxaSEguaxIeazjdrlyUsJ+JJZM1400aZEsSybiEtLVOWBvpzVY0oVWIRmNN+0EbQt++JJoPUMBYYcO8KQTw3tUgU6imUG//SL6Jg4dpinJByEq2zoH26s9gEKh4AaYGUuylCTilKQbOupz1XhTEpAkqKZaYqdIURgynZlSnDksSCarxpwQMG+z5upIaxwHBrPMaA7CfWgDB9atiyoxnGPfYRvMKwmo7jsiAllZIASdwETZMU9TeyarOjQY4OpiUWKvhsdQM8/MESbhpRgLRCuuj4CnhsylIzlUHyNy4SLFoP2K2u5Bl2DZe/Nr1UUqgFGqNS4hrNTR4r0mNjgWEsSlQhxO+ADQEuvy7+9OZCeOnjVUbLVIXMp6VVvy8D3iTXLxuRRkLaVk7dZVp6x1sXXvp4FUycF2TqffDhzHveF5aLNZ4z9TWpUYVolMv0WGO0Mk0LVmC8DX8g3d1UHcKQ3hplR4HUMBYYlIo9ffpvRrdh1zzsmF2zudFIOSaBctuVlViokaIEz8PRN3vZJrj91c+U6zGtFiOBOgkLbQCSvvTQ8fuJC9Bpde96GffYRyXPfB9gh8l4AIa6IWptaRINQaVpooyPgmTBBEFkuYZLFKtCcBKzuJAaxgKjbw3rb1n9wto0TJ8aRKcz3InIDE1JXEhU4AGu3b7918la9YTrMDk0UEaBjCQbt+5UKq8cYUMTgm2pzTuaUZ1uas2ImlF8avpyK+XaGIuspqFKLJA2FkaMNzU+lVocp7ykHSR2LRsLXhwLRB2mxbRJXHRMXEUdJ0zeWMoD18Mokvrpz+esqIf3P1/PVPOlCTa6GttW37hXrC4BMHcyTmkXSkEUuyUMmqtxlPE5THh3scqjyV2StHKnh7FApClRhYQCjuTBvitEgTAOhIxFhMZhNhEljRbjIQVgUzSNay92onvf/sIYLUmGHSYY17HLNm5H2sIk7/CQNkhJLCjxJqhlStLAWstVx6io/tK9LFlID2OBOL6UdD5DYsF0NzWuCjFTjgmy8mUkyRdaBdFJXNjSL7MlM9IkbSWJAZ8TLtAs2BwPvOvY8zBmvJmcayC0oDNuN1DUfG2lUp/S9oTCgcDw/CkahRLPNNQpYiPsZNtYaDEWt99+O3ieB1dddZUhctSBWT7CrmxMGwuWxCJiVUhadOORfkVEfAUr/oGVwlWSSKr5yhF/eF1s6DmgW3upMmX6cdnGbcI0xm43NVNMWePNz9ZKpafbWOCMN2UlREaumRc4EJTTGFBmLKZNmwb3338/DBs2zCQ9VoE1fmG7myLrQdJjZDBK1MctJ+GjFktePPKK6CE1duImNib0qq2GoZRYBjzINOv2phazBXKL2UU7EQG2KoTmFcLPo05E/h+6el2iCMof+fKCZYvSxwklxmLLli1w3nnnwQMPPABdunQxTZMaNI03g2DZWOB6D7sIiBiVWNxNo6/SKKJS5djsG5QqRPg+YStNDDj1gL7SeeQutMKUZwYJCM6bWMgwFvgAWXIwI7Hg108+e2fBOrjuXx8H3idJja3EWIwbNw5OPfVUGDNmjDBtY2MjNDQ0BP6zAVzEyLAqhD4YorKx2FUMu6KkLyobC1v3xgDgmBY5gUWyx0eSINWuERpXJ32OxwnW+KZttAWJRQRxJ2RrIOOqiLr7T29+DrOWbZKsITpIMxZPPPEEfPTRRzBhwgRU+gkTJkBdXV3hv379+kkTiQHOeDP4N8tYJjqvEP57nBGfYaKSxPaGMHPpV3DsHW/B6/PWMNNERf5PHp8Z+NtkP2DKwlyS5SAPmWYz0U9YpEViYYNBUpFYYAJQyYBWrmwZZAwaUeTNMHY0tcBrn7DXxaghxVgsW7YMrrzySnj00UehpqYGlWf8+PFQX19f+G/ZsmVKhIqgYrz5/KyVMO6xj2BHU0ugI1mTWOYeAwyMxbEwUkryceHD02DJhm3wP/+YDgAAb3yanIlk1HYz6godlIDpAmOeX67DpcEN6S3I2yy51rNSmz3osGl61OA9MSYgFXlzxowZsHbtWjj44IMLz1paWmDSpEnw5z//GRobGyGbzQbyVFdXQ3V1tRlqOcB0YHNLcJb/b1tI56F96wKDMKpLyEzwKaYoyn9zcuUVANsag8Zyv3np05I0SabfJIQeRQxp3K4ElUW9qQXPCWDWA3Pz01BBMcPGZ7DKpBtv4ihobpFkLKgBrdQhdjcNomFHk0Zt5iHFWJxwwgkwZ86cwLMLL7wQBg8eDD/72c9KmIqkgbWRf7VtJ3TrUFX4u2FHMzWd6cktDOmNNhY1QY09RElfVJE3wzAax8KwUaDs6SstUOmT52etRKdFGW8aGhju4j82WE1DWwuKXiH89pRt76LthhkUJCu+D+s2N0LP2pqykllJMRadOnWCoUOHBp516NABunXrVvI8amCM9ninEUynRa0KwdZmQkxaDoO25EbRBC22JkXVONsaY9U5KCLK8ZeW/rbxHTK3m+aXcNHaa0IVovOt+bH1q//Mg4c/WAK/O+dA6N+VHZMlaZLaFEXeFKdhMQYe4AaBeVVIgnQhbUiw7SYKZU4+AGCNAvXe7wqwPZYj9QpxPcqGhPGmDz68s2AdXPTwdG6RsodI8x6Drf8+/MESAAC4/RX+TcxJGx3at5u+/fbbBsiIBjwXQczENc1tiy2TzZQTVRnMsi0Ne2qpMXEWUZ8okySt2VWB2UxMbThpCeltA0wbC8qxOecDXPDgVGGZ8jYW+X/JfObmaMYrrzmfIomFeEfR7RfTsQtMRd40iXK/3TQuRD3lJ3JcbgHKaxEqV0RqvGmonLhh59p0hiRaw3jThMRCTxUS/Dub8cpqDKSHsdDMHI8qhP8+yoFkU9RqRqKCKyQNbBHmUxsRF2A52AVOFeKMN22DtY7SVCFYNCuKiMwxksGSMp5XVnY2qWEsogBaNYEsT3xtevQjqZxsLGjNkwqvkF3EGDdOmNjwcV4h2tW0FWSonJhhY0ljX1muXr8sI0eVWEiVEMobypzJlJedjbaNRVKgs5944JkVHUdqG5Hs63NNIS6GQQa+78O0JV/FTYZDG3gjJiopmqmp6SQWbLAjb9LS4tpR1cbCFMLF6Uhf4kBqJBZl1u4AINbjRaoKaavMRjPa+g4aB296HFRXYKeID699sga+df+H2nUm3Ri3XJBvgl99fUjJu6jsm8ypQowUk0qw3U3VFwN5iUUbLUQ2E+6meWQ8jzvgkjbf08NYSGyJvMtpooS5kN4JG1URoLklB8s2bi95bpox6lCNE+r5PsB/5602Uueu15t2ccHhe8Dw/p0Dz0y0cbTGm+kYFTa+QipAFrJM6TgWpt1NQ38L+IrEjY/0MBYSO0pFSEYm6jRZvP/FelQ6E9eml4MK14bY+bGp0cTGl2FUTHnUJO30Ua7g9Ybpa65t1oOta1cFq41pqpAnp+HuqpL1CqGG4NZYVcPlLVq3FeasqFcuL2qkh7GQSJuljTiD2LQNF7fd3LXpRooBgPJRKbFsGUzbYmCL8yXS2kYrHW4nIhHuGhNzBiWxMNQNaXEftmO8SX9OU4V8tnozqkxpGwvK9WZ631qaWRQkK0lIDWMhg0pK5JQkqkKwQbtMkF6OAbJoML+340r0fd9Y3brtlZI9yCqiaiNT9TgbCzZYTBctQBYW8nEs6M9VDzrlPofTw1hIdGA2G1KFQDw6qiQuFuXgfQEQXbwKmeYw1XTOeNM8wuPahIoiSk+NtEgsbMD0pg4gHwyRrgpRR7n3dmoYCylVCOfWuyghc/U1P125D0NzMM0XYYtLUg+UCW8YK4wYbyJiKBlTdxopJQkw/yVMiYXGPJA13syv5WFSVPu/3Jf09DAWyEGUzXiJWXhF16Y37BDbapiTtNgbyTYmCbsPzXYu1mXN95MVDr3M1yXjCPeMGXdTcRmmpJJJlG4mBaym0ZmPLZybsLE06DCVSfPykEV6GAtkulZphV2vECxEiwXNnZKG8h6COMT1jfGoQnaFHo0XZow3EfUYGrlpGRPRGm+qlykrsch/WKmbqLOxKGtgO5DpERJDTyYpml6CSNFCXNIo3/cTY2Ph++npT1Mg+8YzdFMkKvKmMa8QM+WkETKXkGGhGiDLFMq9v1PDWGBBU4V4EM/NcaaM9IxGI0/oiA4vEawlwzRfEUco3WT2QHrggZmNACexMINyF43bBDuOhc4lZLLGm2b7p9z7OzWMBXYIZTzKJhXTKTdREou4CTCE2CQWrbUbKUt3XLSq9tLSo+oISClCfVNuEou02FjYWPJk7grBokUyjgUrpLcqCbLtlKCtBADSxFgge7Aim0lVSG/fB6NcgZ2JH2Uci3gCZIGfLHfTckEcjKDnedFJLIzdFbILDQpJMBkLDc5C2t2U+mzXZe/Tw1ggN5SM55WkdXEsWhfAT1c1wO8nLoibFCqwTWXc3VTGeNNQnQkaFumBF/xp5Gr6KO8KScmgsLHOMq9N15FYxK0KKfP+Tg1jgV3Vs5nkXEJmCqYm6yl3v2uknDDsuJtGc9TFMqw+JMsrpJzHs220Gm/ql4MpwlzYftehLLBVIVHaWOT/9UueqaDcZR2pYSywQ6gik0lQtAEz2BXXnKiMN9F3hRjshF2pPyOLoBr6O6rIm05iUcTO5hw8NX258XJlLiHDQj6kN0Nqoli/s7EoM1Djx3vxeIUkCTa/P8q2NX4JmVRaQ7ebmhDTG6Cj3MEbC1GFTXfGm0X8ddIXVsplNY3OfBQFL8TQoCexKG+khrFAR970POqCkzSOTwZlTHrigWVUzKpCzJSzq4OUIoU9RMxILDBpnPFmHpMXbbRSro2Q3rKMXCGkt3qVAZS76is9jAWSO7V9ZXrUyPk+XPev2drllPk4tga8KsScWN/E6bRcFqZYLr0zZWOBElno12OwmFhhizmycQmZLK2sS8hUSSj3/q6ImwBT0LkrJC6vEBNYVb8DVtXviJsMLqysJxHtR3GwoeU6FlVgkwEiN5bSOBb65WPKMLWZlgujyMIl/5gOX27YZqVstsQiSsaiTWLhB5+pdpu0jYVaNdawyzEWGY9xCVnSeiZilPvClYd5d1OsKsQ3dvpOSVeg4Jly0RDWQ/wGM8xbtNemR1aVFUyct8Za2TbuCjEhsdBDeXf4LqcKqciWxrEAKPduTDaiPIHHdW26SaSFycMgjvb1PFMhvaPzCkmDjYUtsNYXnQBZ8jYWpc90eky2uz9d1aBRm3mkhrHAIkuRWCTlGnUHPBqb5K41VoWUjYUz3pRGVJ9qw3gzUlWIkVLSCVYT68xHWeZ+wdrN8Lf3FkNT6Lp1Z2NR5kCrQhhc7K50SqSh3L7+pTmrqM+Nh/SWCZBl7K4Q/TJ28eEsRHQBsvTrAYhOYjFs9zr4eHl9JHWZAlsVouNuKpf+pY9XwUsfr4Kj9+lRfKjjblrm83eXk1gA0G7K9Mq+I5OMqAzlAOIN6W0Ku5LxZnQBssxfQoZShZjyColgSDz1w9GwV4+O9isyDDvupmoNPnPpV+qVEvB9XzqWRpKQGsYCazg3amBXehwL0wSVG3b5BtBEwlQhuxJzIgvPA1i3uVG7nGiNN+3Xlc3EY/OiCxvXpqvu6SQz0GrQrVaOD/IXoSUJ6WEskOkOGdC1LCdPOSPSyJumy5PxCjFct4M5hL1Cvvt/U7TLxIWxMOVuaqQYLjzPK0vOgm1jof4xqowcyQz4vnq/+b58WPEkIT2MBXIMeR6UTJ6IPN4SjbR8fqwhvQ1VHZVhYRJQzobTOONNM3VFsce0Lo3l1yGsptG6K0RxApnqJx98x1iUE1hMuRMdW4SxIEHiNKaXRerdMhSY3MjLhSkwAZvfyryozli8EYyNhZkPjELtEksUVANIripEA75ThSQC2CHkAf2ukF0dqfGKMR7HQuaukARdQpaS7kwycHeFsN/96Ni9zBFjAB6UpwSJfW26epmqjFyYGdCxsXDGmwkAelGnSCw8cAuxAx0yC4OpNTltY3GvHh3iJiEU3tsMdBnAwX1q4eD+nXF1RRSdtAz5CvaV5RHYWHSqCUZsCIb0Vq7e2VgkBXiJhUPUSLrxZlUFexpgy/MN3kJWxusJFXv3LD8XRgx0+6lVQoAbNNHYWJTn6mhDFYJlCmprKpXr4NYPvlOFJAF44036XSGyJ4Lbv3GAVPqko3yHcBAqp5SMB1DBkptKlGduYTagCjFAhSnw2iWyyJsWytSVIshICKKxsXCqkDyw0oKwxIJEq6eYGhG+Lx+kK0lIDWMhs3SEO1t2Mn17ZD/4zqj+cpl2YfxrxnIjcQNsojJLnwrYxcnksl/GBxUqePOrnG17dElvtfdC1qVXFRrlKLVgMxb2jTd5EgvfV1eX+QDQXMacRYoYCxxYBkoy3V+OXL0INtf3VfU74NwHJmuXgyFRpWs88KAyS8+JV4Ukzd00ORs2r12iMqS2UY1uP8nQFEV3spjrpIM51iO4K6S2nZ1bMVojb1opOhLscneF0NJ5nlxI7zQyFrbx+dotkdSj0jeeB1BhYFF1xpt0cFUhZfyxRmwssF5Hltvp5CG9YZ9eHctybbNyVwiyvTtW81Qh6rjnrc9hTUOypbw8pIexQKcrdTd98eNVsGgdfuNLo7vqrhzHgxIzrQCZxcmcxEK/jCT1JlcVYrVi6k+DRrbRSSxs2lgcs08PuO/8EQBQnocmtvGmTpm4dO2qstz3qqqlaUvM3DkSF9LDWEhILMJJZe+yL8O5V/bAnthUJzJr/LhLyPQRGyNuuxm1y8dbb9r0CilHZoIEO/KmfYlFFU/SWcbSOF2Up1KNAuyGYmIOlftETDPUVCHJ6lAj61GC1rQktK6NPk6LjUVwA05Cb8mBfVeIeploxoLjqr4rQ6pV7r33Xhg2bBjU1tZCbW0tjB49Gl555RVbtEkBPYgMuFSVo+W0CLswc93WmwzjzRiYjnK2O6CB7xVis2KLZYMpGwscPllZr1eZgI7C7zJc2tjXpmtILJCGk9lMhqlyiXMWx72GSDEWu+++O9x+++0wY8YMmD59Ohx//PFwxhlnwCeffGKLPuOQcfFiQUd355BAcJjNOLq6xYiNRXKYkyRMFytxLDTbWIZp/Wz1Zq26+HRYKzoS2LkrBN+3SZRaxH02kbKxOP300wN/33rrrXDvvffC5MmTYciQIUYJk4WcjYXeTEqa6NwEkrMNxQNWj8qF9DYzLqIIhhQl0jhfAExF3jRCiiY8yq/yAdsrRL1M7BT0wYfqiizsaCoVccQ5jeNeQZSNN1taWuDpp5+GrVu3wujRo5npGhsbobGx6DbT0CBnKImFjI1FMiazgw2obGK8HHGoveIWY5pGEqablTmvy1gkoWEguAEnhSYZsOaLDkOLZu59tsQiznncWnd8nSktw5kzZw507NgRqqur4bLLLoNnn30W9t9/f2b6CRMmQF1dXeG/fv36aRHMBFpiob9V6IjYEouU7GVKAbI4/SnqaiuBlwwExkkSb5JeiYWJkN7xt025d0+cxpsAANW8u4ZS6hAlgjRjse+++8KsWbNgypQpcPnll8MFF1wA8+bNY6YfP3481NfXF/5btmyZFsEsoG03DYgsyn0iOpSCNRGFjIVEWizSpwqJmwIAG6e3tPSSF1CFJKKzUOhdWwMAttxN8WlZjEW8xpsxVg4KqpCqqirYe++9AQBgxIgRMG3aNLj77rvh/vvvp6avrq6G6upqPSoRwJ6KjLibGigjabBt7FeR8aA5gisaVSNvMt8JejvjeQVGwNRkTl2ArNjqtVuz7rXWJgzJTSBD7ItJoAeD1685GpZu3AYXPTzdSoAsrBrDB4CqCn6QrDgQt/G2tjlrLpcL2FAkHTI3CrKQcW4h0ojqHgIlxgLUjTdtLMSps7FIwHQhadi8o7nkPU+cbQ0JaBeA8pJS5FGVzRbotuMVgk/LlFjEabxZThKL8ePHwymnnAL9+/eHzZs3w2OPPQZvv/02vPbaa7boQwM/hPRPCeU3DeNHRdYDaLJfj8oi6XkeRxXCL6+1PrOz2MwlZAYIMYRy2Lh279IOLjh8D7jx+ehc5xNjSO5Rf5Yg49mNACoDst3itLHwfT+R7qZxQ4qxWLt2LXz/+9+HVatWQV1dHQwbNgxee+01OPHEE23Rh4ZOSG/5ynQLSB6StBHRgCXP9EItKs7OrZnmy4wTidg8BfA8L3KpRVKMWsmTPY+mbMaDnIkgK6bQRmqc16b7vrOxoEGKsfjb3/5miw5toI03QX9Cl8MJTBbWx2FEA13JK4T3Lg6vEBMSiwRZWSRk/xQi6o2+VQUXf+Pg107z0jkd5One2UJ3o9KLvIn7zqP36QFLNmyjvovV3bTcbSySArTxphF3U80CEgjbngjJWY5KwTfeFOQlLeoNjYu4TxvmEf+EEbUpz87GFpLCcMlIe5ME0Zqvd7upeBL+9pvD4Oh9evDdTdVJ0ELca0h6GAuJdNo2FgmbYHl0EFzhy4PtgRiZC6VC5/BONkIbCwtjQdfbIGnuqkmdL2FEHZ8mMV4hHo45Tlr8HrGaUp1ezAwatUdXAGhVESUNca8AqWEssDAxN5Igviw3aAcTMpyORHsOQyaWWJiHblv5fvwnFhIJXHdL4HnRM0BJ2aexZCSpHzH9pSWxQDD3Sek/GuL2LEsPY4EV57X9TwdJmmAkErSXlEDXINHmt9VUchgL4eJlIfCSblslbCDExYjLdk30EouEIOAVoi69Sxp0wgLIMPesZtmV7wpJDWOBvivEhDK1zCZYEhAVB63SNe24KiSR9SY6JRq6EoucH7fpVhCxhTUONAK/RWJRSyRkGcGqQpK07LXayolcwdWBOQjl609QsxQQ9+EiPYyFRO9q8xWa+XdFRGZioZCnnYbEgnxt6hO1VSGG6DCFcpkv0XuFeImQAmApSJotgdhjy67EIl88q55Y2XvHWJgBWgdvQJeagLVAGnEvCknb7EjwGAsRbERh1Y5j4UP8RxYCcW2eUocNz3mFAIgCZCWEYMB58URxbXqeFt0yTCNumWV6GItIbSySM8Gw+NXXh8Bx+/aIrf64jYl4qDFkvGnqE3XbKmleIXFBlmGMw8YiCSsJWhUSAS1SsGj/JCOx4CGumRj3EpAaxkIGaQ3pLRpMfFsCu4gqmqTK6VhLFUIkMHVKMGHomiTWIi4+/KIjBxZ+Yxba6L1CkrGSYMkY0K29XUIkgKFZh7GQcvlmVNOc82Pb4OOe/9K3myYVMsabu6IqpBxpDgA5UzIeQP+u7WHpRno0PBq4jIXwdtPib1PMkwnjzSQhDgnf+FMGQ8dqueUtam1hHC6udJASCzpBv/jaYGhq8eGjpZsiokkMofGmjrspihHNG2/SK7rjtfnqBGgibglxaiQWUUaPS8pJQxYJ22+s4e7vHCSVXk+SQ0gsDLVvs+Z9DEnr5zhmi+wU9Yy4i0nWGXmNdJAMFY2ewb07waVH7xUZPRhgvHh0GFrMxpwvPYnbQdxLwK7HWBiwsbA1kGwOUBfUi40aXkheoSqk+NvUKWH+ms1a+RNmu5nIhZeGOCQWSUDZhvQWvM9o7G5SXiHq1VhD3PM/NYwF2b3XnbwvO5UJVYiloaRbKk/Hnxyxq33ISpSqNWwsskSCpKgg4haDhhGHhC88R0Ut4kEcdCbF3RTpFpIwiO8K0ZBYYOpPcGM5rxBDIMfQbp3bsdMZqMvWyaYcvU0ig0TTyG6slVlPmcMn3XgZlyxGjtaQ3slhLsplVO+qEougKoRNVJI2Upzxpnr5MsabSenHAJzEIlq0ntz1RoKtmBA2GYskjn0pICeKyuLHvYRMorzESCziXlXCiGHwydtYxOAVAsmYl0mQmqjAZoAsGS8iGwzXkL61WvnjXgFSw1iQXcsfUCauTbekCrFpY1Gea4cSZBeUbIZjCIawscjnTQpjkcvFs7AcsXc36vMknXR5KBc6baJsQnoDJkCWOsEoG4v8v4bbpa5dJVRoHl7jXorSw1gQ/cDrExMnE1sSC1264hxMGQ/g4P6d4yNAA9mMuioEoLiAJYWxiIsK1sacpA2JhVgkFl4Cr02PkQ5ZiL1C1MtGzeW8xMJwo/m+byCWjbOxMAJyUePrCfVhI4wzgG1ViP0lo7oivgBcOuCrQvjwvOIClhwbi3gC87CasVw2q+jvCkkG8B51CYJX+D9OEh2JBYYEey2iyxjEfcZJT4Asj/67NB1/OGQzntBwJ2tpASp3481yJZ8ngRJtNsWLpPzEGEzmfB+enbkibjIKiGNch/tN1DceeDEZb8Y/aYJqZE66+ElF4bqT94VzRvSD7TtblMuQm8vJa5i4V6IUSSzov2npeJsFRs2RtdRqViduBGO/XBaeMHiMIuaTihKLuKdzK5pafJizoj7yelnzivY4iWMlahuLpNh0kBLYpNAkAi9AVt+6dtCjU7XWGMPM5YLxpoUm0z2jxH3ISQ1jQULEdfPGAcZoxprxppVS7ZddrMNuLbYmi053tqpCWgtoSYjEoikmnQzG/nVAt/Zw1ZhB1sejbPmkSisqJCW2DJaEpDEdNqnBqULs0aFtY+GMN80geFrinUD5BlMYNYc1d1NdS2BDdKgiCYukCviqEH5eD4qMRdyTOY+4JCdsz5rii1+fMRSuGrNPNATJokzHrzaQauSkQUZCZrN+lfo6t6/kvo9b4qCL1DAWJMR6Qo4qJBsjY2HTeLOcVoyIwWUsEPnzTZsUVUhzXIwF4nm+qW2Px3DxmMibmPl38+n7K9NUUqdAehoVsF4hSVpCktB2OvXbtjuKmy9JDWMRMN6USBtGBSLAvFOFMOpI0sojAa5XiMh40/MSpwppjksVgjhBJk2cTgJD2Q+OGAiXHWPmQq6ktEUyqChfqAXls0AIAeduaggBd1PuRsEvB2OYKSOxGLNfL3RaW5KQqFCu5Ou2ez57UsSXurejqoItsSDnZry0sDN4aFWkqW9IjI2Fx/ojuWg1wheksfwpOsabosOpvvGmXn5dpIexQEosRO6mpiUWMoOuUtfdhDOYbE8yH+yefGzOE604FkT+pKhCmnJxSSzEz4thkJMHvBGjofoS0ghlGyArZmrz9atQIdpDdIPtxb0SpYaxIMG1sRC8x5xedcOtslDFub67HGBTFbKlsVnbUpoFbp8jOAuvwFiYo0kHcUksWI0VNKuOZjMojWMhSE/Jwy5bkaiSOj1qe/SurTFTAZqO8gMmaql11ayGvZB1VYhzNzUDrDSv1eiHY7yJimMhIbFAp2y9ZdMWojgdJWWBkqVDd5InThWSMK+QQJwEDfGxbeAjUJohPjFt4JX2j4MYem7q7Mw+6Esc4l6J0sNYIA3EhO6mmDgWMoxFlKoQHh0RbPtJWZRkJxWvP0XtRkrAkmK8qYKD+3eGCd84QKsMTPdHFYVTthoyHonpspnlMMqK2vAOfW16UiY4IL21rFOhDtE+o3tIiXspShFjETSyYKkrRAZTGDWHrZDeNhmLaJDkqcxGti0ktwoCXiEJsbFQQfeO1TByQBetMmRsLJKIXdXGIm5bBVXYaD8ZNXc+pZrxJv+9/koS71qUnrtCQr8rsxlozpXGiveAP5EwpxaEfacSqmxKLKJQhSRkfZIlg3d6wKwzSQuQFUa7yixsb+Lfm2Ci75i3m5JeIYFn9hosTImpOBatiY2xFnSJRcTjCMv4JWR6A4BYpS2LMfv1hG8f0h+2NDbB1U/ORtKQN95UsLGwzFnEvRaV+xG5iMDk8Nj2Ch7A2s07mMVUYAJkyXiFSAy6yookTV15lKu7aeuGokZ8QBWSUIkFtl9090ucxCK5gwRvYxFtfbZRrvPWpLtpdWUWTty/F7SrxJ+19SQWAlWIfJFG8+siPYwFgYzH9rDwwINpS75i5jVuvBmhjUXcQVHKVaTa2p/qbZe0AFkq8ECduSqUweblhWmMQ7IiGYbHqI0Fpc2jHkXkt/OGcFIYoTxs0CPDZOnULxRYOBuLZCAobvWUN2mMNKIsjTcjWBWSsvDUVGal0vOaHXMqSppXSBhRuVEyVSFB86fQDz6OGtRdiyYZoI03jXmFJGTCEODFT0gStZg+kOmnfEoV42KVdrEvsXDupkYQ1hOyNmlx5E27qpA9e3RgprVpY0GjxXj5SVh5PIB9enWUkirxA2Txy/GgaLyZUE0IzoLeM7BxMCUWxReyC/e3RvYzSQo3fdQmFl7h/+JFJiCxSOggpoA1N3U+QaZvCwGyLPShi7yZEHih3ywbC9EYQNlYSMnLgn/+/cJR0LGarsezGscC7HKxvp8cVYjnefDrM4ag0/PvCsHU1/pvUm0sogKrqXQib0blnipTl0kbiyTMGPKzeUM4URIWz6yNRf7bVKIqy7ZLdUUGBvXqxE2ju1Y7xsIQvBBnwZZY8AcByitEYiCF01ZkPeapQNvGIm79aILWHZmJpXVtukd6hSSUsUAyR7obB/sSMnWJhaphYbgaUd+I3NADNBmydqQx4lUVmei9QojfuqGko4SN5UZlCshkeejCQ2Dq9WOgb2dOdFXfgMTCqULMIGxjwTbe5AMVx0Ij8mY24zG7vLLcQ3rHTYAidE/FURtvSgd/MpxONr9OuVEekuPwjggzY63q0KgDZBGqEE66RAksJBhBGdi6ByqP/l3bQ127SqF09+enDAYAgN06t5OvBJzEwgo8D+Cm0+micLGNhbhJZAQL4fqynsfsdKtxLCLY9qMUW4sgQwqfUSx9F2Y+i6oQfJ06sHJSY8RUkCqDZWNBU4Wg7RnUiJId7zJeMcZsLCjlxHFfUFAVUj4SCxFUuknppmOVLII8pw3rC9N/OQZuPH1/+cITgNQwFmHL8xEDulBDFIsWHMzeLsXVhv7m3Z5a9neFJIevkAKvz7fvbC55Rvb/jqaWyFUhthg4XeYTI7HA3B5MwlaUWxrQ8T4Msnbhkiqz7INHFIgzOJdUPuL/TULJeFPB+wSTo3vHauUvjJs/TA9jQf5uGx1UCYCgpzALn1wci2DabNZj6r8SZRylAB3qR2iGk9YBb6Om6tOJR0s2bCtEYo3KeNPKMDEgWsbYWNRUZtqqsysdkM4nYWMSp8Ri7JBeZionEKdXiFU1mRST0Ao14018PXnYlu46GwtTIDqK1+Em3E11JBY8VYjuxsTVj0JwMZ/367FadVHr0JgscbJUvD7fv08tXHviPnDREQMLz8LJ8+3KsrE4/7AB+kRS6kOnj4hhZUosiBftJGOMRHZpGchILIL46dh9oWuHKvk6vVL1E08deujArjCgG9tdXRVorxDjNav3r+fZUe6qrO0ydBTCgCMzqe4IZSWxmDBhAhxyyCHQqVMn6NmzJ5x55pkwf/58W7RJwaP8pjIWgnJMR94MV8jLa/vES3Kx7avMXxOjM9HjtM8Q1f3jEwbBmcP7Fv4OL2nFAFmleS87Zi+49Og9tWkM1GfhOGCk9RmFNLUUG6ZakrHQE5XL5lGTWOzVowOM2a+nQn2l4HmG2ZoiZLFR21jofJOIYVa6w0NmaVcgXoUZUUHcljJSS9Q777wD48aNg8mTJ8PEiROhqakJTjrpJNi6dast+tCg+8qXdp9oMBgP6R2ioYLjFWKTsUi8u6lh+mTWR4z7IMl8lEgs8l4hlP6zYb1uyxBXWxXCoKuxuXgBWl5igXbtjEpiIdFP4e/MZjJKfUKLY/GbM4cy1wdbbUGOf+68sVB/0tS/Kq7ESp9gWxUSs8hC6tj66quvBv5++OGHoWfPnjBjxgw4+uijjRKmB7a4SdSdxt1NQ0kzGY/JTtp1V/Sse4bolB/nRUioMO4BVVvx9/87aR9487O1AMBgLMD84mmjrVrF8noFs7LvaCq6y8gaKKuLyu3lCW8+MldtB+oLzZd/X34419Yo47Fj4OggVomFRj5W3rxkVsoQsy2tmipEfj+wvd6VlcQijPr6egAA6Nq1KzNNY2MjNDQ0BP6zgUAcC06neR7AX88fwXyP4VjlQnqXgmVYk9O1sRAsCrYNenQmi+nTmExxPNVCvklJZpIse3DvWqFXiOk1hDVG/3HRKK1ydelk5SclFrLMS1QHWjI0uwjh5s9k1Fx1W6UkxYx5+wrmOLLVFkTBUd8VojPv7cSxkK9fqe85rUn2gCqfV1Y2FiRyuRxcddVVcMQRR8DQoUOZ6SZMmAB1dXWF//r1U4v9L0LY3bT1GUUVAh4cN5itDzV9CdkpB/QuecbqdEw4cVUk3d00Tokops9J1TeZOpPhB8iyoQqhLca/PXsYHLZnN2p6TP0eMh23DEb+xqbSAB/YqtRtLOyJtEui6aoyFoz6WXuCxzH81gG5nJWNu6lnViWYL0nOK8QL5MXVw5amm0WZeoWMGzcO5s6dC0888QQ33fjx46G+vr7w37Jly1Sr5ILsJ16Hex5/I8HwDFhVyLH79oAj9+6BSgsAcP5he6DTJhF6qpD4OAtMSO+AjQWRPkNY9tMCZHkWVFAscuNUJwHgbCxkYWpciDZMz8PXFT5YtI4fNc5CblOys12Q/ca/K8R83TYlFiol216GivZ/WKj1eNwSCyXXgCuuuAJefPFFmDRpEuy+++7ctNXV1VBdXa1EnAzIAZGf96xBwleVmFOF7N+nlloXq887t6+EDlVZ2LpTfSFmIYo9J2F2WGhkMuKTYMDGgniezXhcVUhUEgsffO1FWjtAlozEAq12iG5QYWsKM3CmbCxk6zUFsonLJY6FLXsxtcib9qRjyqoQtWzGICWx8H0frrjiCnj22WfhzTffhIEDB9qiSxpeYOH3Av8G09EXtb+ePwKe+uFoVF1Yd7+w1XdB1MnRoVqLqhjBAq1TRZLdTQGCC07QQ8QrBsiKyMaCRm7Ot6fK0M2/TYNRVt1MGyXjq7PWBRrC40XHxkJmcFjzCiGK5dtYWKgfUeRB/TrDrWex1e3MohXaS6WN1SQjdte7uCUWUozFuHHj4JFHHoHHHnsMOnXqBKtXr4bVq1fD9u3bbdGnBJ5RDWtyjNmvF4wayDZCJSFlvEmkzefjupNpjLe4uVQdxG1jwao/P0EzDOPNjFeUWLC8Qlh9qn5zJ0Vi4bc+Vy4TTDBAeFUI9mSsugDvUGBmsFWF539FRm3LZeVhNY2tORKVKuS2s0qvWMBs5Hd8cxicMrRPCS1xG2+StESRRwZxu5tKMRb33nsv1NfXw7HHHgt9+vQp/Pfkk0/aos84ROoRTIdjxWVh7XpRZE5Pn/HsOYR6YE98mIfOWI5Ti5LJiGknuzzALGaKbpqbd5TeKwIe28ZCSewKdMY27/FDW6hZtZQkNSyx6N6xGn7xtcFw5KBSOyOsA5Qqo7S9KchYiDyiZCxhwt+p2o9eaGwIbQYsGW+SH27T3ZQWrlzP4JufWVT0u9cdV1KWbUlC0cbC8lpstXQxpGws4uaCePC80t+0rmMvsuKOnvHLMVCRyaAHnyxX7SHpUIHn2Xc31RkepsW8MrTwJFD5V2SaILPI3/w8YI+BbMYLRKXEglZfQbLieYBdVsiUJsZduIRxx+0FFx4xEJpaclBbUxHwWsFuYMoSiyZ5iQXe3TQsscgY8QrJg7XOWrOxCFRupw4AOv2YNmcJcnWbo3vHUts/NVWIvAQbbWMhTU1bvnJShSQZAc6f49KjEwa2W8dqqGtfKUeXxEAlPQxUEPdg0mFcYlWFYCJvslQhGXH8A9Zb2Zs+i/WXltizU3UJbbLl6J6iWL1fmc3AOSP7Qb+u7QvP0IyFIi1hiYWwHolDQHi8ZDNqbSd/8GBfYKgDcvzalFjQVdMaeYUSHvz7/E+V23RNMpWmcN7/TYZN23ZaroWN9DAWFIkFrftEHWpyg5Mtysvondx5WVu5frvDWSe+l2lJDba4OTefBLTLoMIgFxxy7c0K7Bp4RoGqJ9AwP3LFcXvDifv3aiuTMuYZ9Xuh37pdILMnYcPXq9KkIrFQvd00y5BYiBjW8HwUzU8bd8QABL/H5iVk6uuPnMu2Cm9UlNzJ55XJkk9r21g954OSNNQU0sNYkL+94L9AeRcJwpUJ6ta1sRDlTbIqJC6BRaeaVgkUi/agiqHtGfG+9TlHlcJZEis4F07xQNJy4v694P+N3bewKapEDiz8rUQNCfwAsH3D/PaQ8SZmbGLXBlqALBqEbqgMET87uR0biyBjEbHEAi2yKM1nwz06qrtC0KoQrcOael5dpIaxIFF0N6W8E7S2VXUCwkBQ6+pxnq1ABFu3jg1OnJMAA/K0SH5ma+RNdj6euFvV6I8VUyP8ToTAmNAcewB25o7quN1BiZ0hArbtSlUhdCpFjAWrOnbkTTFtKggwzRa9QujjS1woaw4JVR2CsoPq81bYDjKHaUNTtoxxLqmpYSxodzmoLJQmT/XSqhDQs7EQqnkSbIkcBeOjA3J8kRM/i7KxoL9X0ecChGw8wnYStAWYTRguHRK2T9My6FQjF/vPk5AWhjcf0jMo/JxbJzBUuByvMduwKbGgNQf2k8LJTMSzpUoslGws5Jl5bB49u7X41lSlyJtJRCUhVvZC/8rA5Lwq6VcRh+3pccxxn/qjvhkxSvBUITzdt1f4v1KYkFiE65a6eTfwO9mMHRY/Pn5vmLOiHq47ebB0XuymEl6wWW1eqajqYtdrtDiiXKTEQlf1oGDzxk0jPknhX7f9EVWgvijC78c5o1PDWJAXePFsLJKMVhsLDVVIwIGw5GWibSxMwzQtQXGxH3jOVUFZUYWQ5XuhdxKMBUVvrQPb4wuDk/bvDdeetG/Jc+FdIYD/flqALGo6oSqEPt+5AfQsABt5Uxc6NhYlHkwGmoI2b20HyCrsTcj0zsYiZgRPB3lxk3w5JmN1qNwFoDUYeHkjWPO1VCEJZwIDqhDieWscCzbxGU5URswdE9eeuE9pmRwbC1XxpwljuCQwllqqRGRempSIllcksSjtO369nmcnlhA28qbuEZiqCkHZWNBTCW0oZCQWbbAdx6KQJ4IFL05VSGoYiypSFVLgCuUb1qa1uoga214h9iNvJmBnaYPpOcW6WrrVxoKdryrLDqiGofHwvbtT8hGqkBKJBb6ekvGgLbEwj6jWRhnGiqYKoeUVSywkT7uW5q8XGNs25zBFFYKWWJSWpG9MWvrMtiokX3oU49pJLAyAZmMRN2Q7tnWh0VCFCLJaV4VYLT1ehMN450HeFUID7x4JbNTBMMiDcHjvUlWFmFiok+QVogKsGDysCmFJLMTepl6oD/gZshk7cwx/CZkepAyLQ2lU7DOE7wOSv9bfKrFCpOaNpCpEB3HugyliLIhBIqvIIhCrV4in6xUSL0uVZndTEkHjSX6fVVawwz0r3zEBlLFeoE2mHP7fskiCjQUL4rtC8Ew9LaQ3NZ2quyljHqnGPREDqQrRhEzwtmAaln2G+UUjabeb6qmXnSpEG6TEIqPOV9hVhSAIshV5EyREvapIkCZEGjKkk+t7ViCxqMxkmO1Om/i9aoP3F9D1wGQZ4jJZCJzYTAyNBLib6nwHlikrlRLR+wnnTownuDJjP0AW18RCc5CQuX9y/N7wxrXHqJfFoSXqO4tU1FmRqELsV8FEKhkLWV9hW1CpXk9iwUEEm76ORXnc0hYZkBuGKEBWZQVboqF+KVNQFRPILzGjSyQWmvOljPnKNmkhUmJBdNxundsxJY0YGwva32Q77ta5XeF3ZTZj/a4QntRRd4aSY3PobnWwV4+OqPWOFd9HRtWBhdKarWK8GcF6F5XrLLXu2Go2jEqau6lCOWbjWAQpQIUVNld95ChniYWUmjS0sfNtLNhTjLbxYBYDMltJsCZqfkaZIf2+tirEhseCrMQi4gX7z98d3lov1XVRwFgwNkyyGU/Yr2fhty1VCEmCVXdTigpPde/zNPLyYN140wv+K0K5qpfTw1hUlH6KSsPG7dnA47JPHtIbvn5gX+Z77qSIYJAlga9Q/UwZ2slmzgpOupVZto0FLR9GtRGIMhv6YrmQ3vy6ZZGE/leFzKfz4ogE0glWV9xpvQjy8GQSQa8QK1W0VUR7hJHQsYw3RYwbHvk2ULF7klOFBP9NK1ITIMuUu6nNiYVaSDhp7vrOQVBTmYUXZq9kZFajyxTKWWKhCqHxZpYtC6DtE9KqkEz4nTA7tRwTp5skeIVEcUrLBLyC2uqlpBPZWHihfFRZE1EGT/qlA5JMuyG9xYw0C1S1kRf82wTpUUTEbK3HfkVOYmEAARsLDTGbUa8QJX0dG6peBFEhbmlPHGgNkMV+z5NY0BaXcFkixYbJyJu6SHIcC2HkTYl6ggHK2JyF2CtE7uMqK+wYb1Zls4Xf3GvTNftCxETxIFYrUZ6pSBKU7DJk5hyHE22DqS6O024tNYxFIKR36F8ZkBP3xR8fCcfu20OLLlnwJhDmBMREwiNvxg3VKSi6hKwiKxfHAmdjUXpilsmfR5BB0V+IksBYRiKxCJ2UAehth5JYUJKQhxvyfaUliUV1ZbHcvXp0tFIHAENigcwbPlTpuubzIHt+UyGDzHP5sXvBXd8+iJrOhfSOGTRViFociyKG7lan5/4ZIgCrT2TB9AnINMr5EjJVykXGm7zIm7R9AtOFZL6SMUbTYzPKND1eTPb+Xd8+CGbfeFKEZy58Tdg1QTaOBd3zgVCFZD1jbbxnjw6F39WEfdrPTxkMF4weAM/86PBSWnQlFjSGDKX6a/23OmRHx5Pctb5XI9immoImGanMZuDM4bsZr8t5hRgA1d00bqMDBdiMY2EbZcxXoHD6gX2he8dqOG1Y0YBW6BWSzTCbHuMVQivalMQimDdZc6Vf1/ZQ175SOl/UXiE8K39h5E2PcQmZT6Yp/jZ1W+oPj94Tbjxt/8Lf1RVFVUhdu0r41RlD4eD+XYzURUJXXVHbLjgebB2kRAzhkL61cO6ofkpl0zQhvNoG9VKXIMU5o1PEWJhpxvDmqFNq1Axj3KoQHYlFkqM25vGnc4fDlF+cAHXEApfNeMBb7ys4xp39urQveSarCimxsZCJY4FgYqSQgABZLJCkXXLUQK16yDbO9wUtu7xNFF9VUJk1Y2OxR/cOgTFUU4kbNNrXpiuqQvL56sKMhWQ5WIi67aWfHAUTvjFMrw7kPB3Stw7+dsFIpTqcKsQASHfT5lwOABSNN43ebhr6G0GP3l0hyTp1yqBcpB3hS8dEt5tWMlQh447bC86iiD9LxOOUJTNYf/A9Ta/Poo4XwRMA4NxR/dGbDkApc2hGPyw3pqOYAtkAY8dOp3oXDOu9Ka+QsG0HKbGwCdF4EyHMWAgNOiXq8DhzSiavMC1Fmi7Kf8J+vQq//+fIUqaYTZdThWiDNGxqbmld0cpxm7VGcwyqkGP2wRu+mmYsbDIqgTgSnscVndJOrX3qauCnYwdT88lLLILvZBaTcF1VIdFLn7oadFkAhoPLWcynenKm5eeFaFb14iKbMeBuakgq63lB74+w7QIvn169lHbHMF9t/5ZILCytabKMhcq4V6VdpKZJCtLDWBCTriixkO+E8BjRGbwqeXXqi3vMhU+sMt9SDqqQPEquzeZKLFrffXTDiSXvaAsS5lAaPF2x34kQdJtsXbSeJYz2ZIeTjai1NjYP3SJNeYWEaaEbbxbRyviZaeSdzbnC72oJqZQOAsabhZM7Pl+nmgrqcxOQkSBoVsStOy1IDWNBng76tsXX5w2QRy4+lPrcpGeDyoDRs+SNd4Dq2KeYljDYXBxol1CxkA/D3LVDFbLsYGF0L4/iw5L0qFrYZddUZrnveYjzZuBCPlZGg3EsgpFP2fl5B5vRe3ZjVxAQWRR/mgrp7YEHjc0thb/DkipboLWGTLu3qwyqbFjrq+4olJU0qahc4vTYiAKpibzpeR58fPNJ0NziQ/uq1s/idd2Rg7rDqcP6wEsfrwo8j1vXb228Jd54Mzr8dOy+sGLTdjj3kP5K+cMLD088Wcl5R7PnwUjZgjp+dT0+T6WighLG0kCZSVx+ecazJCo4ff+bs4YK8wMEN88KQ8ab4AE0NhUlFjoMS127Sqjf3oSrVtElNJ+mJsRYiL1uUGSVIO1RMaNAaiQWAAC1NZWBk6GS8aZBetRUIeojLu7BqrPoRcnQ9ehUDbeddQAcsHudUn4ZVYjs6QeTPBDHwrDqLCCultU1q5PCpUMyp2Ku1nw/P2WwOC1VFVKKrKRNRD41K0BWVTYD+/buJFUmq56dLTlhupJ8oU6ZePXRUgEEReNNhDBjYc3d1OI6WpBwRaV6iQmpkVjQIegxykoYd/RAa4qQCAavXtNF1+66J5IwI8GTWPBOg7QvJmk7YDc64xNUhbDfiUANTa0Bs8abeRuLaI0sDuOpKNpAU4XQdgeexAILsoSKjAffO2wAbNvZAkfu3R3OuOd95XJJGwsVWgAABvXqJBlCXpHpa8sWVoUIJRYSY1qHoVYBr4q4peYmkGrGIm6JRRgyhkpK5cctsSgx3sQTFOVkkpH80uwGZIJS1dbITTGy7Gd/dDh8tnozJQ3bxkIGdGPB4ElKpl9M3r7JkwRg8oUhsv+QqY+mCqFKLEwwFiEbi8psBsYdt7dmmUEbCxF4Y0DO1ZJOCxbtqkKRN8tRYkEZLykUWKRLFRKGqMOoi03JI52FWz6vnsQi3iGqpQoxR4YQuhKLktgRjFn0xKWHcccArb16dKou/K5gXGDGW5Rkvsy0jcU5I3eHobvV6hdkGbx5Iht7gpca5RUiOCmTtJpi3DwAaGoxM+PkIr3Svk+MfJqwKsQWMP2mioIqRFV6Y44Uq0g3Y6HkbhqvHMpaSO8IoNNyYVcym5C7qKs0bVj1wSpPhbEdtUdXuPKEQfDHc4cz82UEm1EpHfQ0omusZRnVdpUV8OKPj5LKI4LsmFadAgWJBaIAWoAsWj5e3AkeE84K6W0sQJbXGvysZ6dq+MHhe2iVJXO6p3vO4PPLMhaq66HsvqFSDc1OJ01IN2OhkMdmkB/egM0IFra/XzRKur6oEfYKkaGnU00F/PPiUXBgv87cdCLxMmaS6kosjhrUHQAAetfWaJUX1hkDtDItV5+4D3z9wNb7SGibO2m3YYoRpYuplYvWRpIXW/rJu/QZRhUSUD21/cvwNuVKLPbvUwvPjztCWB9Aa9t27VAFk8efADd/fQgqTz5fHnkVn4y6h8yfZ6pR/dyWpn1VRBFCLe6Kqiq+ckOqGQsRaEyE2SA/Mmm9wL8khvStRUWxtKFzvPWsoXAc0vJb193wqEE9+P79YEZvrVtEr9oamPHLMfD2T4/l0iQaSgf16wzfGrl7QFqDIa2dRqwJEqIIlB7oSaG0QnpTQh+j8jEahCUFKKlPkjHNR7CkSixM7FCkPQ1n4Hqe/IleJ4rjez8/vq1efhnfHLF74TetL2X6t0Yy9LjM15GfEY27abpZi1QzFmrGm+XrFcKF4md1bV8F7ZAnhVKPGvzX+JwFmoQJ/afMgsoaD906VhcWcmx5V54wCAAAbjmjGMPgt988EH58PGGMh/g+8v6O5py4Y1lF0lQqSTEellFNmAQqpDqxavK8yFASC0oSskzyNW/sex6+rVRtsch8tTWt4bVFn3jK0N7F/JS0h+3ZFV0vdh3ShV0bizzDTDxTrK9z+0pp4/CokG7GQmECxXXfQX6C0hY2LE1WPPMkytQ63bb9K1qoTLjw0dtYnXosSVefuA/M+/VYGLN/r8Bzmji88DelbFJioeI2mIfJS5xMQ7U7lEmWYGTIdsuTScsmO1ZFNggmpHWselQhM4Zoaa85cV90Xf27lt4GTEN+Lstt2MQctDju82UbEWZBtEbvMkgmu2MIogFCVYVIlsGvX0L/KCGKZdfHrcA6TEReFDGDskGHqGXIXC2OaDiZE04+KiwLmBNzFXFpVJNCoKNiXbSnJJPjxbZy2ZQc0j7bK/yLkFiQjEUbmbRxiZFkiVKQ9MR9ARVtaIoZC/6G3a4qC906VMGGrTuF9fbr2h7+dsHIksvITKNrhyr4Yt1Wq3WY8OBrbdtkshapllioILbgJAZEv7G7mxoY5CeGTvNhmBBTytyyiPkm1oI/pC/O9ZJnIU4/xRYfhiUWIvdUVjkiujBIgtYYQzMvDUpiEVg1W8cHbVxiJBai+gISC1F/CWszDxH94Yvu8pBZZ8l8J+zXC0buIVafhPPJIH/XVOAZ56ZfJVWGqc5KJl+RbsZCbQ8yeJFSeJPgpS38SxHTK9ZnCthFIBc6PKv43R/YrzO8fs3RcMTedCNOE+Jg2gKtowoJl3fT6fvDnJtPgk418ierMGU0ssjqShgLiRWLektnErgDAlHRk98cUCHVaRILSka+8Sblnpi2vmN5hZhThaiVQ8slklhgXKOTtjf2qStlLHiGsSprR9DGQjp7sW71rFaRbsZCsMjSTqNxSSySurDLgGy635w5VMpKnWz3vXt2YqoMsHprXj+atvoO7x9VFRklpgIAGaCJGNcqdz7I1CUDk5bu6jYWupuXnCokx2EsMCo3mTYzxlgYKaUVMmG1WZ8q2ph1LtpTQd/ObOmELgo2FhyC43YgMIF0MxYKg00nFkNJ/RJp8wNNZ3G24yaFL/OkNjVG37oa+N5hA7TYadZaY8LGwrSfeunV5Vr6LHESUmJhiLEwaa1uEtLBihRJLkgMURKL4u/8JkBjeLOKA40c+6TTjylvBeU2otlYCDgL8rXcvSLopEbKINPS4svw80p8F8WWTme9iPtuKxac8WacBJBJ2/6lzVPs4Il7Gxh33N6wd8+OqIucxKB/M3Zx5SUzzYCFT5LStgkMPTQzPfG7CeEVwlr4hPp9BC22YYdVZpeKan/PA89rXT8GdO0AAPQNFmVjESi39D150OHxKXHZV0kZbxLPRUtaxvOgJe/doUqcIkxJhngwYrxpgA5bSDdjoeJuaoGOAjjk8C4zMoIIGNuqigyc3hYxUrdKpsTChI2F4YXDJKOCiqNAGm+aklgUVHFi0TULJvm1qA9iGBE1iU9+NRaaWvxCbAUaE2HCi4M8VJhThaiVQ2sbEUk0O54w6Lf8AuCvSWPZIql9J7WdDY1tbZV3KF8y5RW7kCrk+6MHwAtXiEPeIuIN4etXSFsp4wupU2EEkBPT+Zy/ijARzdD0gaRUFSIH0am1JD2pCtGJY4Foyrj1veS3/vLU/ZTLQUv9kJ3Xvqoi4PZI23RR9kCCJOR6JGJ6unWs5r4vVKk4/o/ZtwcM7t0JziGiaQqNN4k2UL2jJ2qNXDQSC+K3THWhYZxQTYg8YzFp0iQ4/fTToW/fvuB5Hjz33HMWyDKPb43sB8N27xx4liBNSGGk/fLU/WG3zu3ghtP2l69P9FJhvnhePAalrI3AxCnQtO2AripEJ6+pAFle6N/w7yQAo2JTdTctfr/aV9MuHONdQoZFCzEPRIxK1w5V8M+LxXcKqW5G1RVZeOXKo+COcw4sPBO6LJO/A7YpfHps3tdBA0knTd1qai4UxpkRt3ntIqxBuvu2bt0KBx54INxzzz026DEKlXaPyxgmv8j379Ye3v/58XDxkQOlyxAOVsVPU20SE5E4wzARedN0yF4dIRNAcIHAXdtdTBM2NparF68fR5WnTEkpCpdURczeqA4NWr+hrk0XfF9OUhVy1CDxvT4mx4zQKySgCsE3Ls2wGAvdFZzWzqZ2BVrofKmv85h/JArSS+Ipp5wCv/nNb+Css86yQY9RBAc1Lo+J6JGFvBIdb8QKOvT33d85SL/QmMBa+0xILIy7WYZa3vZm6AHALWcOhe4dq+DWsw4ofYmEKI4FGeEzatDujiF///DoPan5cNfIs/OpDg2aii68QZ08pHhvBs0dmVZ38PK05G0k4rlUKhUzU649yKpCVEhNXk+ahXXjzcbGRmhsbCz83dDQYLtKAnyulx7S216ALG5azrswnZ3bV8KmbU3C+s44aDe48olZeCJiRPgbWb1gYkIaF7OGiZIkMiguxkgsAM4/bAB879D+xt2TyXnSviqbWB1uXXu5OCH4IHP09uxQlYXbzx7GzEeTWpEb1OXH7gXXjd0XXp27GrY0NkOv2pq2+vj05EwafbXBZIlGjDcpgyzOgznt8GKKhHzROkbSJOK2gWLB+pFkwoQJUFdXV/ivX79+tqsswITEQgdfbQvGv+eRILNBvHXtsfDEpYeVlpEwPlinLVkqKbKZnvnR4Uplm5dYmEN4TaPqnw3EPAnXRStLdK9JGEa9QvJlkuUjWlqVBJqNCYk5N48NeDyFQYtZQTIWNRVZ8DwPTjmgD5wzEr8GWuArjKp7aeMm/90H9usceC8172SZc4Njz4S6lQUTtwiTSCrjb52xGD9+PNTX1xf+W7Zsme0qC2AZDhVhN/LmkvX4i2xkxlmXDlVw6MDSePkJlJRahaqthG2rb9nSg77+OImFibrpd6YUf7eP6JpqLKIY31QpjidWwdEkFhgPL5G0SsceIgrQ2mu3zu3g45tPgmcuDzL+AeNN4rvo7qZmTvRYkHVQjTcNExFg6hXZ4SSv99ZVIdXV1VBdjXODMg3ZBRugVLSk2umdqivg+6P3oL7LqzL27tkRPl+7pYRWEU3lAh2qMQuqquTBNF8R7judRQjl1aBcehDBxa0U7aqy8Y88GydXjlsI7RVmnNEkFiYkYzaMyU0WSeOdPA+gts2GBCVxokrlyHzRIgp3U95XyfRP7POTgXTHsSB/R6wKmX3TSTB0t7ogPW1E/Ouyw+E7h/SDh35wiDR9WkiwkRG23bF9uk+vTsx3tAVfp9t12yjoFSKZQSO5iGmWl1iYGy35DZUVz0B2ngaMIDnpaO8wfSK63ZTZB4KyW2wwFga3I/ElZGqSh2g29yLIcRZJHAtTVSSUs5BmLLZs2QKzZs2CWbNmAQDA4sWLYdasWbB06VLTtGlD5NJDN940A57odO+eHeH2s4dBv67tC890xOe8ZwFEPAh1TluYrPSDZ+vDQ/boCn86dzj1hlXbFud6pduzIwgjcMimFNq+Mr7AvLTux9lYKIqVCyILyjvEeKEZBMvea0NLbcPGwiTELsvF33J3hXjU31FA1nhThTqVQ6+JeqOCNGMxffp0GD58OAwfPhwAAK655hoYPnw43HjjjcaJ04VS9LaYbzelQeU0X24In6JQjIXgi08/sC/s36e25DntRKLTdiWRN2XF95y8tNOlKcZItGC3i8DG4oyD2AaRAAptqdk0KmGrARASC0Y+4Q3MCVeFyBzuZbomYoFFALLGmyrNaWIOe17p+rB7l9Ir3+OA9JHk2GOPTeyNamEE+07NxiIqGLmURlSEiipEYwLoBchieYUU6VF1G6XGC1ArilmeKnABsuTesdLTghA1NhdvaJBVhai0g4wBLiap6oJdiONBeYehkX5tur7KKqceWJUJk1IQusuyOG0g8qYgbdQ8hmlVKQ02hDDtq7JWPVpkkGobCxLYzSQunsnEeOAOVh+il8bIGCGF0mIWP3XjTbOTL1yadPGSi6gxVQiloG07i4yF7PXRSjQwBn5+M5f9VsxmzmfM1FRn9NtNEV4hgqKxXiFxeQhQ7+tiuZgyaKQdVLXmqOY6J80QKtRhKjRAUs/4uw5jgUwX7qeoJizfKwRZRlkrQ0JAqULUYN/dVL18lFeIIfJphnUDunUovs94UtJJFbJoXTFmv15wQMjwGQvdvqXlxrQ3TapBekyo9lk5xrEIvqf/Fperlk8VZB1RnPqD32dItWmkFDNI97XpCp0Xl5rHxNjiluFB5CNPR63EVIWQvzXF3sagWR6ZHaUKkayQlZ5WVV27Spg8/gSoqYzmzEFjBP7vgpGF30GXcbXyAIJ8Kq/9qDYWiI2GrgqRjGNBoctGHAuzkTfxY1HOeFOFGjMwcW2ALQTsBinjJSkCjFRLLESLEo2JsNkxvMnSQTLKIbV87RKiKRMD1noqe5KhFWP6ErLwBNfaDCLUhbD02L3raqBz+yph/r9fNIp7Hw3GkEzmdkwMVPu20GWU7Ko2FkF3UzW6rATIMlhkr1p+jCKEJoQKnUvIVEDWEIXEwpQNichWJS6kW2JB/kb2Xlw6qw7VbH02WoqSOHdTibSCv2nAnIC6UDZI2+6mTS1yFneyiy+PfplFWHf9PGafHjBvZfHun/zm+cjFh8LHKzbBifv3EpZhmsnDGPTKe5qoSiz09VpWVCEGF4LjB/eEy47ZCw7YrQ7GPfYRALClMKw4JCLjzagRRd1xSmSiQMolFsRvpNgo/Kw6ohseO1QnU2IRF5h3hYAcp3/bNw4oeWZa1BleJHa2qC/cOFWIGZgIm0zbyI8c1B1+dOzeyhsyCVm6MIzKoJ4dpcrEDBdav5EnX8zhgOoVklTrvDZ4ngc/P2UwnDqsD/U9JnQ17ROjtrEgQb023XA/cMMLSJSRVA/NdDMWSnEsgh31s1MGw57dO8BNp+9vgB42OppgLETfiGiDf148Ct649hhtWnRhSmKxW+dScbzxkN6hv5uaJSUWkuPU1EKL2fhF/aArpja9aTAZFWJenzy0N/z6jCHw94tGoejBhfTmSyyaGMymqOSkx7EQQZVBiFNiIa0KUaBVaW8qKSOkCkkQj5FuVYhCh+3XpxZmL68v/N2nrh28+f+ONUcUAzyJBd4rRABEQUcN6mHsqma9200ZLwycZEx7hQzv3wX27NEBFq1rvXROVhVCQvsSMqnFW65ecRnyEEkYpA1VUXFAPPj+6D2gfnsTpb5SqAbI6lRTvNq9th39mnehu2nC41hQITkQaKqZqNmKYHycaFUhJmtLCnORcokF8ZvSe7RO+MWp+8GlR+8JL/74SGt00WBGYpEeZQhmfiiL7yUyYiZqVUUGXr+6KOXRsbHAXJsuv9nSn5uK/qcDIZMX8ZBWCpUPdJVQVTYDr111NPzk+L3hrOG7ieumPItTFTJs91aX37NHiGknESSZdTLnf1ecEgsakzi4d2kEXx2Qw75GI15M4A6cBC3/qZZYBE+3uFavramEX3xtP0sEsWGCsaDhqEHd4bPVm2H0nt3gXzOWW6mDBRkjsZL1k2ljQfxOirspBE85LLE3ChGqQgLSBlVxrCYxqvlpodrl6y7+5o1VjEEoNUx8BmDf3p1g3977smkQdLgdd1NcmY9dchh8vHwTHDqwm1T5zYSYRVUqFquNBeWOl1vOHApdO1TBt0b2M1RLsQ5TofOTIq0ASDljIWvoFye6dOC49yEHDO3w94+LRkFLzocK2v3GDMQ5qfNASSwUyzYdRjoMaYlF4DdOlG8CJsoRSQVFEA3LcJmvXHkUfLqqAY7dt4dUPTrqRFV3U+nmoBpvyhYiBnYD6lhdAYfv1V26/GaCsUbZ8VDo0XE31fV6ofV31w5VcMuZQ7XKJUFWUV2hxlgkWUKdalUICaoqJHoySnDNifvAwf07w3dH9dcuizYBPc+TYipMwqSNxTkjdodO1RXw3UOL7ZSUkN5h6NhYoDwUJctsj4iRwipT1Ie6bSnL5O3Xpxa+cfDuxpmiPCoUb8OlfQcqFLjIxsLQMXRg92JEVdvrXjOCGxKOqxh3pkiuTSd+R3HZX9RINWMhcjeNGrTF8CcnDIJnfnREogYXdtH+xsF83avJS8h++81hMPPGE6F7x2JAHnUXSduMhdyXByVEesabJO757sGwT6+OzCBWDTtKjRdloctYiPrCZk/R2rq6Igt//u5wuPNbBxLpxGXRNiMUQyK4Wh0rseCV8s0Ru8PDFx5SfGBZZt5CEF2FcNenUWPCa0IGB/YrhpCP5K4Q4qNqJEIaiEMVJeG4nHbGgvwdP19hH5a+kbUO3fmtg4zVEb5Jk1ZnWPISRZ+qrME7dbxCUBIL3IefOqwP/PfqY2CfXp2o77/aulOuYhotisx75/atXhInDBYH0Yoapw3rC984ePfC35iNhmaAh2JIBCL/OoY3iQx+d86BgTtgbG89pMRut87t4HuH9YdLj94TxWTkYZr3H96/M/X5xKuPht+ePQzOOLB4SJIN2qbSnuT3yRwqeetRUpgKgLTbWAgGSFKDi6giabwTtnkP6tcZrhoziJtX9dbJOKATx0I3WKNMk2wkGQtFqHbBu9cdB2saGmFvQbAqm3pkbMmYcVZTmYV/XDQKnpu1Ap75aEVr+YqSDhK/PHU/WNuwA84fPQBFKwa2l72WkJjlN2eWBqkTQfaOGBEuPWpP8DwPLntkRuD5oF6dYFCI8Y7E3ZT4Kh2vkKRil5FY0HBw/y7G6vr35aONlRWGaB3IL85XjdnHeN0663o3nkEqgefGHQHdOgbvHLBpvEmDycU276anBj29PEZq8I3hu0Fl1oPzR+8hUSujPsUB0qmmUshUANhWhRB/cPof+41H79MDjt23ZzEfgnoRY9Grtgae/OFoOG1YXxQNGNg+UGFsLAKgGm+aoYXEyUN7UwPm6UJNFVL83U6CseAZ1idB3Z9HyiUW/PeXH7sXdKqpkLYwp2HEgK5ierRroeO2sw6AwX06QW2NvtjUJK47eV9Y1bADJi1YJ50Xs/glTWIx8eqjYfKiDXCurCGutBeO3nf//lsHwm3fOADmrqjXKgfAjMtq0iGzyZHjFhVYK3BRmQxV6rAtpw1LLEQ00ET4wXDzcg3Tr2t75rskjtHq8E3Cih2UJFVIyiUW/FFUU5mF/zlqT9i7J10HXS7wPEAxFZWKxwDVyditYzX8gxI22RRMLhKssmTqGNSrE5w/eg8tLxzUXSGa3+15HtRUZoPiZmSZXdoHx5ntU5KpPtYJNKYaUA3nFZLAnU4TGMZCBJV2eeLSw+CWM4cquchGDdIORUZiEcZFRww0QY5xpJqxsC0ZygfpOWxPsbRCB6ZEl9edPBj6d20P10sEAItLvIb5ZJP6d1Z9UZjhkF+BuaeAl6JfV7yoV7b5aiozJdepR6COJmC2Muz3y+jcyVMjpvworugOI2mmZTR6hg/oLF3OYXt2g/MP49uiJIWPayTssFRtLDwPYPzXBgeeJaVvU60KsY2HLzoEnvloBZwzYndxYrA3qLE3sPbt3A4mXXdcyfOuHaqgJefDDafpX7RmChixXlIWCZOgxVEIg8dQ/fzk/aCxOQdnHywekypBiEqYXMt9EKQrnlVTde/HhQI3a6SIQZJvTH3j2mNgxpKv4LQD+8D97ywCADPMq+xaMWqPrjB1yUb9ihnoQMSWqdSQcOrktQnHWGigZ6cauOyYvdDpTc/nHx+/Nyz/ajscsJuaseDvzzkQHnx/Mdx//gjYrXM75kIYxzqEqTMtYmSy3SsQkYF4C21d+0q0G7BK64W7xXYfJKGLTd8tQ4Lm2liZ9fTCwpcx9urREfbq0WrUO/OGEyGb9YxKJrEM9KOXHApPT18Ov3h2jrG6SfTv1h5+9fUh0K0jzsAdgyTxi+lmLBLU0Dpgfca1J7HvIMDg7BG7w9lIaUvUwF2bbp2MSCCtCjH03RkFG4sSWsyQYr026g2aHvmeDZlxJmteEDTebP39xKWj4ZfPzYWbTk+OBNE0yE2Q1WTcaw4sozKbgf4cI1ASqnPngsP3UMtYBkg3Y5EwJOH0VS7A2JUkyb3KFDDBmEx9t8p4DHdLWqRGPMicmGXtoWj2GyMGdIFXrjxKqhwHHNI0XMPrQJK+LZkKGgtIUJs7IGDz2vSkgfwOjI2FqcEcjJqJQ/jkTzIWSdbd04D3CsGXKdsCslEedXDxkQNhj27t4duHmLqh0wySGqgw6etLmD7fT446xEksygCxDhbK5KqtqYARA7rYrZf45h8evSc1SdInvgowNhamPlv61kjawuUF35uGzT7Gli11d4RkG2QUmDtV3HDa/vDLU/dL9K2YtlHuX14u9O8yEotkoFyGBR8zbjgRHvzBIeKEGiDX5/EM99i0qELIDRlnY2Hmu0keBltkqSqEeKdPkjWImB7e50sZb0q2gqmbNG8/exgqXRKZij+eOxwAAG5MkFcaQHyrNW8MJXmOkUi1xIKMaNahOtWfGimicHFC2Vgkb41UAhlQSHTbJYA5o1UTjBm5UdkQadtkHrGOrHI2FnI0mGAsKjIe7NcWU6dcQG6epw3rCycM7hXJDc8yfTm8fxdoV5mFPYgr58XlRydhDn+JiQvrTCHVu21NZRae+uFoaMn5Zc1YJClUa1TAWNenhK8I3K2AC5BlSGJBiuGRC26YeTAtsXjk4kMDf9tVhcRvY2HEMycFE0GXqbDBgLarysKsm06ESoF6MkrJKa2mf1w0Cn772mfwv2cPg0v/MYOSInqU726LxKiBdqNiOtgBi5ki97UkinVV0JIrRuEL21hQQ1FbMN7EotTEwpzEYuhutXDkoOSFY7YaxyItPtPQekX6ik3b4yaDAY/4fzyqK8QMT9wHv6P36QFH76N/35VJOBuLCFGO+2BsesYYJRbYW1lNoUVaYmEGKozZ7885EABar/NuLaP4zorxpqFy6HeF4FAONhZJwIM/OARG7dEVnrj0sLhJiQ3p6U09OMYiQqguvDL5rj2x9er0n508WJDSDO773gjo3L4S/s65bGxw79ZL3vbpJb4mWwa09R77jIfpvxyjRpAiSFUI6l4KYzYW8jh87+6w8NZT4H+OavXUwQaZwkBoYGl41UbfFSKjCpFshKDLb3lvS/v27gRPXTYaDtuzmzCtaSYUzdBZaGKy36KUoiZZYpt6VUiS0LWDfeOaH58wCM4Z2Q9619VYrwsA4OShvWHskF7cQf7QhYfAI5O/hO8dNgBGT3gTVa7qJWQ/HasWjZQU40c9YXm3QdLUu6aCUqnq90njXbIMO+6mdsvHQE5iIYco41jsyti3d3nfYF1ucBKLCPDwhYfA8P6d4S/nHRxJfSaZCswmK0rTp64d/HTsYOhT1w4uP3Yv1A2cKvr6N689Bob3txxfwwKaOYzFfr1r4Zh9egTUM+ZUIeI0ndvzmeFyCZDVt3PpmMN6tMioK2THrQ1VyF3fPsh4meWKd356LPz78tEwsM2zwzYbFyWbmGSW1DEWEeDYfXvCsz86AvbuqcY1J3i9lsbPTh4Mk356nDAd65N5m6GOZXmcYsUcV2Lhwd8vGgXXnLRP4dmePcyolAIiXMYyRXppjNmvF6WMInTHqcgOolONuoD1trOGwslDesNj/3OoOHGYBotDI2MhQtaZw3eDkbYD2GkiqiVtQLcOMGJAdAb8d3+nNSZHkm6KjgNOFeIQOTCbuMompSNWjjOsME9ikUf+BttuHaqgR6dqI/VimmvobnUw9foT4JU5q+Gsg3fjlmHDOj6T8eB35xwI23Y2Q6/aoCTujm8Ogz+/9Tk0t/hCb4SetTVw3/kj1GiI+HZTE9jVNCxY+xTbB4hTh/WB4wefDO2qsnDLi/OUyuCNoQD5lE9JSnh0x1g4JBIqmxTK8DGB4NlY5DFs987wwhVHQL8uuBsXMQgYDnKarmenGuZNjF6Qs1DCZcfsBfe980XB0ySMbzJu4D1nZD84Z2Q/WF2/A37z0jz4gaXbIjtKSEpk1UHB202lshZAl/SU51xIA2wG+koI3yCEYywchBjYoxh57oUrjoikTpUJVK43bWIkFgCtzIVJmG4v1TXv56cMhitPGKS8IPeuq4E/f1fPfmkfinHf/559ADw1fXnB0woD2XFra8yW6VSwDhvN4tq6FI6xcBDix8fvDTuaWuCUoX2Mb24sqGxSNsTKA7rhw/mqIi6jR5XbTXnQ+YwowjnT0Ku2GtY0NMI3Dy6Vinz7kP7w7UP6S5Un7RViScr2jYN3gymLNxp38XaIF+XCxDjGwoELDwDaV1XATacPibTev5x3MPzgwalw/al8I6jLj90L7n37CwCgu2aq4qkfjoaPl2+CsUNKDRZNo7klHsaCPC1XGLj/Je4IhCr4z4+PhJ3NOehZa8aTSt4rpPjb5J5xzoh+sGePjoUYMklDXLYA5bIx83DmQX3huVkr4SfHD4qbFCYcY+GQSByyR1eYc/NYod3EeYf2LzIWBleNUQO7RhYOngzpbQuHDuwKUxZvhF61RcNPsrX6GHBRRmp0EoWenaKJ98KCLVVIJuPBIXu46wzCsHFR1+F7tQYEa1cZjdTtzm8dBD89eTDsRnGhTgocY+GQWGCMMck9mSVWzngetCTY6glrY6GDP313ODz8/hL4DiHaJw0vd++iv0glxSI9TnSUvOwwTSG9ywF3fPNA+MkTM+HyY/YyVuaAbh3gvZ8dB53b278KoGN1JWQyHpOpSMoMdHEsDOKP57b6MN90ulkfZrdgs9Gncw10aV8JfepqoLqCPpz/eTE73HgSEIWNRc9ONXDdyYOhf7eiVwnpjWLi9ONGKcA3Dt4dTtq/F9xyBk51GIx+KsdkDOnbelX6KQf0kcqXBOjEJCExqk0qc/aIUldoGvbo3gFeuOJI4222e5f2JUzloW0SzzzzOGz3OuXybzvrADh23x7WPJ9MQ6l377nnHrjjjjtg9erVcOCBB8Kf/vQnGDUq2Yt3FPj6gX3hxP16xWaItiuiMpuBKb8YAxmPvTAfvlfybswkEZeNRfeOVdC/a3vIZjzoauDiNccAA1RVZOCv3x+JTk9KLGTb7x8XjYKJ89bAaQf2lcoXJ359xhCYu6Iejt2np5HyHrhgJLy3cD2csJ+Z8kziHxePgtX1O6BTTSX8a8YyOGs43W06j3NH9YfHpy6FqyleSN89tD9891A5Q+I4Ic1YPPnkk3DNNdfAfffdB4ceeijcddddMHbsWJg/fz707Jm8zo0aaWMqysHYqYohqaChQwL7BxPHwgYqshmYeM3RUJHJGAkc5PgKeZCMhaypTbeO1fCdUeWz2QAAfH/0HkbLq2tXCacOS6bEproiW/Aqu/Roserl1jOHwsVHDoS9etj3RLMNaVXInXfeCZdccglceOGFsP/++8N9990H7du3hwcffNAGfQ4Qj4j5pP17wZ7dO1gzYMyfkPfrUyuVT/UelIcuPAT27NEBHlEI6Wwbw/t3jq3u6oqsMT2/YyzkQapCkmwH5GAfmYwHe/fsmOhbS7GQkljs3LkTZsyYAePHjy88y2QyMGbMGPjwww+NE7erI+O1Wtrr6OZUcf/5I8D37UWz/Ndlo+HhD5bADyWNqPbq0RHu/s5B0KOjXFjr4/btCcfty5aofX/0HvDntz6Ho/fpIVWuCZw7qj9UZDOReaHYgtsW5VFF+JuW/3biEDd+9fUhcOk/Z8AVx+0dKx1SjMX69euhpaUFevUK+vb36tULPvvsM2qexsZGaGxsLPzd0NCgQOauif9efTQ8PX259OZrAp7nWVWD7NmjI/z6jKFKec84CGeoJYOrxgyCw/fuBsP7RX95U0U2A+eWmUibBhOeJbsa6tpXwg8O3wN834cuBuxcHHZtnDSkN8z91Vhp7yTTsF77hAkT4Fe/+pXtalKJvXt2gvFfo9+f4GAWFdlM4o08k4p/X344rNy0XVqt5dCKm78ebfA5h3QjbqYCQNLGonv37pDNZmHNmjWB52vWrIHevXtT84wfPx7q6+sL/y1btkydWgcHh8RhxIAucHoZeSY4ODjYhRRjUVVVBSNGjIA33nij8CyXy8Ebb7wBo0ePpuaprq6G2trawH8ODg4ODg4O6YS0zOSaa66BCy64AEaOHAmjRo2Cu+66C7Zu3QoXXnihDfocHBwcHBwcygjSjMW3v/1tWLduHdx4442wevVqOOigg+DVV18tMeh0cHBwcHBw2PXg+RGHy2toaIC6ujqor693ahEHBwcHB4cyAXb/dneFODg4ODg4OBiDYywcHBwcHBwcjMExFg4ODg4ODg7G4BgLBwcHBwcHB2NwjIWDg4ODg4ODMTjGwsHBwcHBwcEYHGPh4ODg4ODgYAyOsXBwcHBwcHAwBsdYODg4ODg4OBhD5Per5gN9NjQ0RF21g4ODg4ODgyLy+7YoYHfkjMXmzZsBAKBfv35RV+3g4ODg4OCgic2bN0NdXR3zfeR3heRyOVi5ciV06tQJPM8zVm5DQwP069cPli1b5u4gsQjXztHBtXU0cO0cDVw7Rwdbbe37PmzevBn69u0LmQzbkiJyiUUmk4Hdd9/dWvm1tbVu0EYA187RwbV1NHDtHA1cO0cHG23Nk1Tk4Yw3HRwcHBwcHIzBMRYODg4ODg4OxpAaxqK6uhpuuukmqK6ujpuUVMO1c3RwbR0NXDtHA9fO0SHuto7ceNPBwcHBwcEhvUiNxMLBwcHBwcEhfjjGwsHBwcHBwcEYHGPh4ODg4ODgYAyOsXBwcHBwcHAwhtQwFvfccw/sscceUFNTA4ceeihMnTo1bpLKBhMmTIBDDjkEOnXqBD179oQzzzwT5s+fH0izY8cOGDduHHTr1g06duwIZ599NqxZsyaQZunSpXDqqadC+/btoWfPnvDTn/4Umpubo/yUssLtt98OnufBVVddVXjm2tkcVqxYAd/73vegW7du0K5dOzjggANg+vTphfe+78ONN94Iffr0gXbt2sGYMWNg4cKFgTI2btwI5513HtTW1kLnzp3h4osvhi1btkT9KYlFS0sL3HDDDTBw4EBo164d7LXXXnDLLbcE7pJw7ayGSZMmwemnnw59+/YFz/PgueeeC7w31a4ff/wxHHXUUVBTUwP9+vWD3/72t/rE+ynAE0884VdVVfkPPvig/8knn/iXXHKJ37lzZ3/NmjVxk1YWGDt2rP/QQw/5c+fO9WfNmuV/7Wtf8/v37+9v2bKlkOayyy7z+/Xr57/xxhv+9OnT/cMOO8w//PDDC++bm5v9oUOH+mPGjPFnzpzpv/zyy3737t398ePHx/FJicfUqVP9PfbYwx82bJh/5ZVXFp67djaDjRs3+gMGDPB/8IMf+FOmTPEXLVrkv/baa/7nn39eSHP77bf7dXV1/nPPPefPnj3b//rXv+4PHDjQ3759eyHNySef7B944IH+5MmT/Xfffdffe++9/XPPPTeOT0okbr31Vr9bt27+iy++6C9evNh/+umn/Y4dO/p33313IY1rZzW8/PLL/vXXX+8/88wzPgD4zz77bOC9iXatr6/3e/Xq5Z933nn+3Llz/ccff9xv166df//992vRngrGYtSoUf64ceMKf7e0tPh9+/b1J0yYECNV5Yu1a9f6AOC/8847vu/7/qZNm/zKykr/6aefLqT59NNPfQDwP/zwQ9/3WydBJpPxV69eXUhz7733+rW1tX5jY2O0H5BwbN682R80aJA/ceJE/5hjjikwFq6dzeFnP/uZf+SRRzLf53I5v3fv3v4dd9xReLZp0ya/urraf/zxx33f9/158+b5AOBPmzatkOaVV17xPc/zV6xYYY/4MsKpp57qX3TRRYFn3/jGN/zzzjvP933XzqYQZixMtetf/vIXv0uXLoG142c/+5m/7777atFb9qqQnTt3wowZM2DMmDGFZ5lMBsaMGQMffvhhjJSVL+rr6wEAoGvXrgAAMGPGDGhqagq08eDBg6F///6FNv7www/hgAMOgF69ehXSjB07FhoaGuCTTz6JkPrkY9y4cXDqqacG2hPAtbNJvPDCCzBy5Eg455xzoGfPnjB8+HB44IEHCu8XL14Mq1evDrR1XV0dHHrooYG27ty5M4wcObKQZsyYMZDJZGDKlCnRfUyCcfjhh8Mbb7wBCxYsAACA2bNnw3vvvQennHIKALh2tgVT7frhhx/C0UcfDVVVVYU0Y8eOhfnz58NXX32lTF/kl5CZxvr166GlpSWw0AIA9OrVCz777LOYqCpf5HI5uOqqq+CII46AoUOHAgDA6tWroaqqCjp37hxI26tXL1i9enUhDa0P8u8cWvHEE0/ARx99BNOmTSt559rZHBYtWgT33nsvXHPNNfCLX/wCpk2bBj/5yU+gqqoKLrjggkJb0dqSbOuePXsG3ldUVEDXrl1dW7fh5z//OTQ0NMDgwYMhm81CS0sL3HrrrXDeeecBALh2tgRT7bp69WoYOHBgSRn5d126dFGir+wZCwezGDduHMydOxfee++9uElJHZYtWwZXXnklTJw4EWpqauImJ9XI5XIwcuRIuO222wAAYPjw4TB37ly477774IILLoiZuvTgqaeegkcffRQee+wxGDJkCMyaNQuuuuoq6Nu3r2vnXRhlrwrp3r07ZLPZEsv5NWvWQO/evWOiqjxxxRVXwIsvvghvvfVW4Gr73r17w86dO2HTpk2B9GQb9+7dm9oH+XcOraqOtWvXwsEHHwwVFRVQUVEB77zzDvzxj3+EiooK6NWrl2tnQ+jTpw/sv//+gWf77bcfLF26FACKbcVbN3r37g1r164NvG9uboaNGze6tm7DT3/6U/j5z38O3/nOd+CAAw6A888/H66++mqYMGECALh2tgVT7WprPSl7xqKqqgpGjBgBb7zxRuFZLpeDN954A0aPHh0jZeUD3/fhiiuugGeffRbefPPNEtHYiBEjoLKyMtDG8+fPh6VLlxbaePTo0TBnzpzAQJ44cSLU1taWLPC7Kk444QSYM2cOzJo1q/DfyJEj4bzzziv8du1sBkcccUSJy/SCBQtgwIABAAAwcOBA6N27d6CtGxoaYMqUKYG23rRpE8yYMaOQ5s0334RcLgeHHnpoBF+RfGzbtg0ymeA2ks1mIZfLAYBrZ1sw1a6jR4+GSZMmQVNTUyHNxIkTYd9991VWgwBAetxNq6ur/YcfftifN2+ef+mll/qdO3cOWM47sHH55Zf7dXV1/ttvv+2vWrWq8N+2bdsKaS677DK/f//+/ptvvulPnz7dHz16tD969OjC+7wb5EknneTPmjXLf/XVV/0ePXo4N0gBSK8Q33ftbApTp071Kyoq/FtvvdVfuHCh/+ijj/rt27f3H3nkkUKa22+/3e/cubP//PPP+x9//LF/xhlnUN31hg8f7k+ZMsV/7733/EGDBu3ybpAkLrjgAn+33XYruJs+88wzfvfu3f3rrruukMa1sxo2b97sz5w50585c6YPAP6dd97pz5w50//yyy993zfTrps2bfJ79erln3/++f7cuXP9J554wm/fvr1zN83jT3/6k9+/f3+/qqrKHzVqlD958uS4SSobAAD1v4ceeqiQZvv27f6PfvQjv0uXLn779u39s846y1+1alWgnCVLlvinnHKK365dO7979+7+tdde6zc1NUX8NeWFMGPh2tkc/vOf//hDhw71q6ur/cGDB/t//etfA+9zuZx/ww03+L169fKrq6v9E044wZ8/f34gzYYNG/xzzz3X79ixo19bW+tfeOGF/ubNm6P8jESjoaHBv/LKK/3+/fv7NTU1/p577ulff/31AfdF185qeOutt6jr8gUXXOD7vrl2nT17tn/kkUf61dXV/m677ebffvvt2rS7a9MdHBwcHBwcjKHsbSwcHBwcHBwckgPHWDg4ODg4ODgYg2MsHBwcHBwcHIzBMRYODg4ODg4OxuAYCwcHBwcHBwdjcIyFg4ODg4ODgzE4xsLBwcHBwcHBGBxj4eDg4ODg4GAMjrFwcHBwcHBwMAbHWDg4ODg4ODgYg2MsHBwcHBwcHIzBMRYODg4ODg4OxvD/ATY9iQ43loRlAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "np.random.seed(123)\n",
    "x_values = np.random.uniform(low=0, high=(2 * math.pi), size=nsamples)      # this will generate uniform distribution of random ness\n",
    "plt.plot(x_values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "MtZItrEX9qwR"
   },
   "source": [
    "**Obtain output samples with noise** (for estimation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 447
    },
    "executionInfo": {
     "elapsed": 9,
     "status": "ok",
     "timestamp": 1691562277671,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "IqoUnJGM9qXc",
    "outputId": "e9bb12c0-9511-473b-c200-ef912fd4d8cd"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f561c2d3430>]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_values = np.sin(x_values) + (0.05 * np.random.randn(x_values.shape[0]))    # add some noise just to make the model guess a little\n",
    "plt.plot(x_values, y_values, '.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bLzeR98B-Fgn"
   },
   "source": [
    "**split the data across training, validation and test sets**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "executionInfo": {
     "elapsed": 492,
     "status": "ok",
     "timestamp": 1691562638890,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "rCwXgQox-NIC"
   },
   "outputs": [],
   "source": [
    "val_split = int(val_ratio * nsamples)\n",
    "test_split = int(val_split + (test_ratio * nsamples))\n",
    "x_val, x_test, x_train = np.split(x_values, [val_split, test_split])\n",
    "y_val, y_test, y_train = np.split(y_values, [val_split, test_split])\n",
    "\n",
    "# check samples intact\n",
    "assert(x_train.size + x_val.size + x_test.size == nsamples)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "HI7gjeHq-FP2"
   },
   "source": [
    "**show all the different data samples in the distribution**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 447
    },
    "executionInfo": {
     "elapsed": 1408,
     "status": "ok",
     "timestamp": 1691562741826,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "owtT8Kvl-_6W",
    "outputId": "3368b886-b8ae-4432-9d4d-89d50e55270c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f561c24a640>]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(x_train, y_train, 'b.', label=\"Train\")\n",
    "plt.plot(x_test, y_test, 'r.', label=\"Test\")\n",
    "plt.plot(x_val, y_val, 'y.', label=\"Validate\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "F3LTnU_z_z_W"
   },
   "source": [
    "**Create the skeleton for the neural network model**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "executionInfo": {
     "elapsed": 505,
     "status": "ok",
     "timestamp": 1691562931863,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "nvfsk9CQAIUu"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-12-18 15:44:27.914419: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-12-18 15:44:27.940018: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-12-18 15:44:27.940240: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-12-18 15:44:27.941311: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-12-18 15:44:27.941408: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-12-18 15:44:27.941465: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-12-18 15:44:27.982802: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-12-18 15:44:27.982926: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-12-18 15:44:27.982991: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
      "2023-12-18 15:44:27.983049: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 4287 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 3060 Laptop GPU, pci bus id: 0000:01:00.0, compute capability: 8.6\n"
     ]
    }
   ],
   "source": [
    "model = tf.keras.Sequential()\n",
    "model.add(layers.Dense(16, activation='relu', input_shape=(1,)))\n",
    "model.add(layers.Dense(32, activation='relu'))\n",
    "model.add(layers.Dense(32, activation='relu'))\n",
    "model.add(layers.Dense(16, activation='relu'))\n",
    "model.add(layers.Dense(1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 9,
     "status": "ok",
     "timestamp": 1691562940438,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "_ufNlalpAfBu",
    "outputId": "aae1d259-5980-4281-b8aa-946b18ace0d7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 16)                32        \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 32)                544       \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 32)                1056      \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 16)                528       \n",
      "                                                                 \n",
      " dense_4 (Dense)             (None, 1)                 17        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 2177 (8.50 KB)\n",
      "Trainable params: 2177 (8.50 KB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "JUqpfTmaAmqw"
   },
   "source": [
    "**Optimizer and Loss function and metrics**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "executionInfo": {
     "elapsed": 1,
     "status": "ok",
     "timestamp": 1691563034057,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "CFAEzhmiArxN"
   },
   "outputs": [],
   "source": [
    "model.compile(optimizer='rmsprop', loss='mae', metrics=['mae'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "dZyG-bpvA76-"
   },
   "source": [
    "**Train the model**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 36661,
     "status": "ok",
     "timestamp": 1691563180905,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "lZKAXkkjA-Hj",
    "outputId": "8a49519a-47a5-4b58-e184-5f35f4e81eb6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/500\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-12-18 15:44:28.532844: I external/local_tsl/tsl/platform/default/subprocess.cc:304] Start cannot spawn child process: No such file or directory\n",
      "2023-12-18 15:44:28.640450: I external/local_xla/xla/service/service.cc:168] XLA service 0x7f54e1643a70 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
      "2023-12-18 15:44:28.640477: I external/local_xla/xla/service/service.cc:176]   StreamExecutor device (0): NVIDIA GeForce RTX 3060 Laptop GPU, Compute Capability 8.6\n",
      "2023-12-18 15:44:28.644318: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:269] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n",
      "2023-12-18 15:44:28.652014: I external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:454] Loaded cuDNN version 8904\n",
      "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
      "I0000 00:00:1702894468.693137   99067 device_compiler.h:186] Compiled cluster using XLA!  This line is logged at most once for the lifetime of the process.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6/6 [==============================] - 1s 22ms/step - loss: 0.5594 - mae: 0.5594 - val_loss: 0.5497 - val_mae: 0.5497\n",
      "Epoch 2/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.5276 - mae: 0.5276 - val_loss: 0.5149 - val_mae: 0.5149\n",
      "Epoch 3/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.5106 - mae: 0.5106 - val_loss: 0.4934 - val_mae: 0.4934\n",
      "Epoch 4/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.4883 - mae: 0.4883 - val_loss: 0.4761 - val_mae: 0.4761\n",
      "Epoch 5/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.4534 - mae: 0.4534 - val_loss: 0.4288 - val_mae: 0.4288\n",
      "Epoch 6/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.4247 - mae: 0.4247 - val_loss: 0.4213 - val_mae: 0.4213\n",
      "Epoch 7/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.4058 - mae: 0.4058 - val_loss: 0.3928 - val_mae: 0.3928\n",
      "Epoch 8/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.3777 - mae: 0.3777 - val_loss: 0.3600 - val_mae: 0.3600\n",
      "Epoch 9/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.3527 - mae: 0.3527 - val_loss: 0.3349 - val_mae: 0.3349\n",
      "Epoch 10/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.3334 - mae: 0.3334 - val_loss: 0.3276 - val_mae: 0.3276\n",
      "Epoch 11/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.3077 - mae: 0.3077 - val_loss: 0.2995 - val_mae: 0.2995\n",
      "Epoch 12/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.2892 - mae: 0.2892 - val_loss: 0.3172 - val_mae: 0.3172\n",
      "Epoch 13/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.2865 - mae: 0.2865 - val_loss: 0.2589 - val_mae: 0.2589\n",
      "Epoch 14/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.2640 - mae: 0.2640 - val_loss: 0.2437 - val_mae: 0.2437\n",
      "Epoch 15/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.2524 - mae: 0.2524 - val_loss: 0.2298 - val_mae: 0.2298\n",
      "Epoch 16/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.2360 - mae: 0.2360 - val_loss: 0.2187 - val_mae: 0.2187\n",
      "Epoch 17/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.2296 - mae: 0.2296 - val_loss: 0.2117 - val_mae: 0.2117\n",
      "Epoch 18/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.2347 - mae: 0.2347 - val_loss: 0.2356 - val_mae: 0.2356\n",
      "Epoch 19/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.2164 - mae: 0.2164 - val_loss: 0.1915 - val_mae: 0.1915\n",
      "Epoch 20/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.2104 - mae: 0.2104 - val_loss: 0.2171 - val_mae: 0.2171\n",
      "Epoch 21/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.2096 - mae: 0.2096 - val_loss: 0.1860 - val_mae: 0.1860\n",
      "Epoch 22/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.2008 - mae: 0.2008 - val_loss: 0.1928 - val_mae: 0.1928\n",
      "Epoch 23/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.2122 - mae: 0.2122 - val_loss: 0.1997 - val_mae: 0.1997\n",
      "Epoch 24/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.2022 - mae: 0.2022 - val_loss: 0.2011 - val_mae: 0.2011\n",
      "Epoch 25/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.2079 - mae: 0.2079 - val_loss: 0.1884 - val_mae: 0.1884\n",
      "Epoch 26/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.2123 - mae: 0.2123 - val_loss: 0.1869 - val_mae: 0.1869\n",
      "Epoch 27/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1937 - mae: 0.1937 - val_loss: 0.1796 - val_mae: 0.1796\n",
      "Epoch 28/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1959 - mae: 0.1959 - val_loss: 0.2469 - val_mae: 0.2469\n",
      "Epoch 29/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.2010 - mae: 0.2010 - val_loss: 0.1731 - val_mae: 0.1731\n",
      "Epoch 30/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.2018 - mae: 0.2018 - val_loss: 0.1961 - val_mae: 0.1961\n",
      "Epoch 31/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1913 - mae: 0.1913 - val_loss: 0.1667 - val_mae: 0.1667\n",
      "Epoch 32/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.2035 - mae: 0.2035 - val_loss: 0.1849 - val_mae: 0.1849\n",
      "Epoch 33/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1889 - mae: 0.1889 - val_loss: 0.1767 - val_mae: 0.1767\n",
      "Epoch 34/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1919 - mae: 0.1919 - val_loss: 0.1685 - val_mae: 0.1685\n",
      "Epoch 35/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1910 - mae: 0.1910 - val_loss: 0.2073 - val_mae: 0.2073\n",
      "Epoch 36/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1920 - mae: 0.1920 - val_loss: 0.1978 - val_mae: 0.1978\n",
      "Epoch 37/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1897 - mae: 0.1897 - val_loss: 0.1921 - val_mae: 0.1921\n",
      "Epoch 38/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1912 - mae: 0.1912 - val_loss: 0.2056 - val_mae: 0.2056\n",
      "Epoch 39/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1894 - mae: 0.1894 - val_loss: 0.1642 - val_mae: 0.1642\n",
      "Epoch 40/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1851 - mae: 0.1851 - val_loss: 0.1661 - val_mae: 0.1661\n",
      "Epoch 41/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1866 - mae: 0.1866 - val_loss: 0.1685 - val_mae: 0.1685\n",
      "Epoch 42/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1843 - mae: 0.1843 - val_loss: 0.1788 - val_mae: 0.1788\n",
      "Epoch 43/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1872 - mae: 0.1872 - val_loss: 0.1851 - val_mae: 0.1851\n",
      "Epoch 44/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1853 - mae: 0.1853 - val_loss: 0.1568 - val_mae: 0.1568\n",
      "Epoch 45/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1836 - mae: 0.1836 - val_loss: 0.1670 - val_mae: 0.1670\n",
      "Epoch 46/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1840 - mae: 0.1840 - val_loss: 0.1994 - val_mae: 0.1994\n",
      "Epoch 47/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1856 - mae: 0.1856 - val_loss: 0.1551 - val_mae: 0.1551\n",
      "Epoch 48/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1763 - mae: 0.1763 - val_loss: 0.1605 - val_mae: 0.1605\n",
      "Epoch 49/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1806 - mae: 0.1806 - val_loss: 0.1720 - val_mae: 0.1720\n",
      "Epoch 50/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1729 - mae: 0.1729 - val_loss: 0.1527 - val_mae: 0.1527\n",
      "Epoch 51/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1919 - mae: 0.1919 - val_loss: 0.1708 - val_mae: 0.1708\n",
      "Epoch 52/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1784 - mae: 0.1784 - val_loss: 0.1581 - val_mae: 0.1581\n",
      "Epoch 53/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1791 - mae: 0.1791 - val_loss: 0.1777 - val_mae: 0.1777\n",
      "Epoch 54/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1772 - mae: 0.1772 - val_loss: 0.1670 - val_mae: 0.1670\n",
      "Epoch 55/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1763 - mae: 0.1763 - val_loss: 0.1609 - val_mae: 0.1609\n",
      "Epoch 56/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1733 - mae: 0.1733 - val_loss: 0.1639 - val_mae: 0.1639\n",
      "Epoch 57/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1784 - mae: 0.1784 - val_loss: 0.1534 - val_mae: 0.1534\n",
      "Epoch 58/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1745 - mae: 0.1745 - val_loss: 0.1559 - val_mae: 0.1559\n",
      "Epoch 59/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1689 - mae: 0.1689 - val_loss: 0.1582 - val_mae: 0.1582\n",
      "Epoch 60/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1648 - mae: 0.1648 - val_loss: 0.1598 - val_mae: 0.1598\n",
      "Epoch 61/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1709 - mae: 0.1709 - val_loss: 0.1692 - val_mae: 0.1692\n",
      "Epoch 62/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1730 - mae: 0.1730 - val_loss: 0.1463 - val_mae: 0.1463\n",
      "Epoch 63/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1679 - mae: 0.1679 - val_loss: 0.1672 - val_mae: 0.1672\n",
      "Epoch 64/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1610 - mae: 0.1610 - val_loss: 0.1365 - val_mae: 0.1365\n",
      "Epoch 65/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1678 - mae: 0.1678 - val_loss: 0.1674 - val_mae: 0.1674\n",
      "Epoch 66/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1641 - mae: 0.1641 - val_loss: 0.1764 - val_mae: 0.1764\n",
      "Epoch 67/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1725 - mae: 0.1725 - val_loss: 0.1608 - val_mae: 0.1608\n",
      "Epoch 68/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1699 - mae: 0.1699 - val_loss: 0.1488 - val_mae: 0.1488\n",
      "Epoch 69/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1529 - mae: 0.1529 - val_loss: 0.1712 - val_mae: 0.1712\n",
      "Epoch 70/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.1654 - mae: 0.1654 - val_loss: 0.1745 - val_mae: 0.1745\n",
      "Epoch 71/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.1685 - mae: 0.1685 - val_loss: 0.1449 - val_mae: 0.1449\n",
      "Epoch 72/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1633 - mae: 0.1633 - val_loss: 0.1603 - val_mae: 0.1603\n",
      "Epoch 73/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1624 - mae: 0.1624 - val_loss: 0.1547 - val_mae: 0.1547\n",
      "Epoch 74/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.1623 - mae: 0.1623 - val_loss: 0.1622 - val_mae: 0.1622\n",
      "Epoch 75/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1635 - mae: 0.1635 - val_loss: 0.1458 - val_mae: 0.1458\n",
      "Epoch 76/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.1606 - mae: 0.1606 - val_loss: 0.1421 - val_mae: 0.1421\n",
      "Epoch 77/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1551 - mae: 0.1551 - val_loss: 0.1597 - val_mae: 0.1597\n",
      "Epoch 78/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1602 - mae: 0.1602 - val_loss: 0.1709 - val_mae: 0.1709\n",
      "Epoch 79/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1583 - mae: 0.1583 - val_loss: 0.1626 - val_mae: 0.1626\n",
      "Epoch 80/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1620 - mae: 0.1620 - val_loss: 0.1437 - val_mae: 0.1437\n",
      "Epoch 81/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1514 - mae: 0.1514 - val_loss: 0.1375 - val_mae: 0.1375\n",
      "Epoch 82/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1542 - mae: 0.1542 - val_loss: 0.1373 - val_mae: 0.1373\n",
      "Epoch 83/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1504 - mae: 0.1504 - val_loss: 0.1303 - val_mae: 0.1303\n",
      "Epoch 84/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1491 - mae: 0.1491 - val_loss: 0.1469 - val_mae: 0.1469\n",
      "Epoch 85/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1568 - mae: 0.1568 - val_loss: 0.1501 - val_mae: 0.1501\n",
      "Epoch 86/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.1516 - mae: 0.1516 - val_loss: 0.1438 - val_mae: 0.1438\n",
      "Epoch 87/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1509 - mae: 0.1509 - val_loss: 0.1378 - val_mae: 0.1378\n",
      "Epoch 88/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1471 - mae: 0.1471 - val_loss: 0.1416 - val_mae: 0.1416\n",
      "Epoch 89/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1451 - mae: 0.1451 - val_loss: 0.1341 - val_mae: 0.1341\n",
      "Epoch 90/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1520 - mae: 0.1520 - val_loss: 0.1365 - val_mae: 0.1365\n",
      "Epoch 91/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1488 - mae: 0.1488 - val_loss: 0.1415 - val_mae: 0.1415\n",
      "Epoch 92/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1393 - mae: 0.1393 - val_loss: 0.1290 - val_mae: 0.1290\n",
      "Epoch 93/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1412 - mae: 0.1412 - val_loss: 0.1389 - val_mae: 0.1389\n",
      "Epoch 94/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1429 - mae: 0.1429 - val_loss: 0.1622 - val_mae: 0.1622\n",
      "Epoch 95/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1410 - mae: 0.1410 - val_loss: 0.1209 - val_mae: 0.1209\n",
      "Epoch 96/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1388 - mae: 0.1388 - val_loss: 0.1450 - val_mae: 0.1450\n",
      "Epoch 97/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1401 - mae: 0.1401 - val_loss: 0.1325 - val_mae: 0.1325\n",
      "Epoch 98/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1422 - mae: 0.1422 - val_loss: 0.1096 - val_mae: 0.1096\n",
      "Epoch 99/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1342 - mae: 0.1342 - val_loss: 0.1267 - val_mae: 0.1267\n",
      "Epoch 100/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1335 - mae: 0.1335 - val_loss: 0.1236 - val_mae: 0.1236\n",
      "Epoch 101/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1358 - mae: 0.1358 - val_loss: 0.1279 - val_mae: 0.1279\n",
      "Epoch 102/500\n",
      "6/6 [==============================] - 0s 8ms/step - loss: 0.1345 - mae: 0.1345 - val_loss: 0.1073 - val_mae: 0.1073\n",
      "Epoch 103/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1292 - mae: 0.1292 - val_loss: 0.1142 - val_mae: 0.1142\n",
      "Epoch 104/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1357 - mae: 0.1357 - val_loss: 0.1074 - val_mae: 0.1074\n",
      "Epoch 105/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1301 - mae: 0.1301 - val_loss: 0.1137 - val_mae: 0.1137\n",
      "Epoch 106/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1272 - mae: 0.1272 - val_loss: 0.1093 - val_mae: 0.1093\n",
      "Epoch 107/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1208 - mae: 0.1208 - val_loss: 0.1051 - val_mae: 0.1051\n",
      "Epoch 108/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1191 - mae: 0.1191 - val_loss: 0.1437 - val_mae: 0.1437\n",
      "Epoch 109/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1303 - mae: 0.1303 - val_loss: 0.1159 - val_mae: 0.1159\n",
      "Epoch 110/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1231 - mae: 0.1231 - val_loss: 0.1177 - val_mae: 0.1177\n",
      "Epoch 111/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1229 - mae: 0.1229 - val_loss: 0.1128 - val_mae: 0.1128\n",
      "Epoch 112/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1218 - mae: 0.1218 - val_loss: 0.1152 - val_mae: 0.1152\n",
      "Epoch 113/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1189 - mae: 0.1189 - val_loss: 0.1049 - val_mae: 0.1049\n",
      "Epoch 114/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1103 - mae: 0.1103 - val_loss: 0.0922 - val_mae: 0.0922\n",
      "Epoch 115/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1183 - mae: 0.1183 - val_loss: 0.1113 - val_mae: 0.1113\n",
      "Epoch 116/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1184 - mae: 0.1184 - val_loss: 0.1070 - val_mae: 0.1070\n",
      "Epoch 117/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1119 - mae: 0.1119 - val_loss: 0.1145 - val_mae: 0.1145\n",
      "Epoch 118/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1112 - mae: 0.1112 - val_loss: 0.1020 - val_mae: 0.1020\n",
      "Epoch 119/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1109 - mae: 0.1109 - val_loss: 0.1062 - val_mae: 0.1062\n",
      "Epoch 120/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1088 - mae: 0.1088 - val_loss: 0.1185 - val_mae: 0.1185\n",
      "Epoch 121/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1100 - mae: 0.1100 - val_loss: 0.0954 - val_mae: 0.0954\n",
      "Epoch 122/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1079 - mae: 0.1079 - val_loss: 0.1159 - val_mae: 0.1159\n",
      "Epoch 123/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1018 - mae: 0.1018 - val_loss: 0.0957 - val_mae: 0.0957\n",
      "Epoch 124/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.1080 - mae: 0.1080 - val_loss: 0.1012 - val_mae: 0.1012\n",
      "Epoch 125/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1005 - mae: 0.1005 - val_loss: 0.1018 - val_mae: 0.1018\n",
      "Epoch 126/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.1046 - mae: 0.1046 - val_loss: 0.0973 - val_mae: 0.0973\n",
      "Epoch 127/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.1006 - mae: 0.1006 - val_loss: 0.1316 - val_mae: 0.1316\n",
      "Epoch 128/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0962 - mae: 0.0962 - val_loss: 0.0861 - val_mae: 0.0861\n",
      "Epoch 129/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0990 - mae: 0.0990 - val_loss: 0.1035 - val_mae: 0.1035\n",
      "Epoch 130/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0988 - mae: 0.0988 - val_loss: 0.1040 - val_mae: 0.1040\n",
      "Epoch 131/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0971 - mae: 0.0971 - val_loss: 0.0850 - val_mae: 0.0850\n",
      "Epoch 132/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0888 - mae: 0.0888 - val_loss: 0.0781 - val_mae: 0.0781\n",
      "Epoch 133/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0911 - mae: 0.0911 - val_loss: 0.1020 - val_mae: 0.1020\n",
      "Epoch 134/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0958 - mae: 0.0958 - val_loss: 0.0879 - val_mae: 0.0879\n",
      "Epoch 135/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0912 - mae: 0.0912 - val_loss: 0.0925 - val_mae: 0.0925\n",
      "Epoch 136/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0963 - mae: 0.0963 - val_loss: 0.0872 - val_mae: 0.0872\n",
      "Epoch 137/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0796 - mae: 0.0796 - val_loss: 0.0835 - val_mae: 0.0835\n",
      "Epoch 138/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0947 - mae: 0.0947 - val_loss: 0.1037 - val_mae: 0.1037\n",
      "Epoch 139/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0931 - mae: 0.0931 - val_loss: 0.0918 - val_mae: 0.0918\n",
      "Epoch 140/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0759 - mae: 0.0759 - val_loss: 0.0798 - val_mae: 0.0798\n",
      "Epoch 141/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0958 - mae: 0.0958 - val_loss: 0.0912 - val_mae: 0.0912\n",
      "Epoch 142/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0871 - mae: 0.0871 - val_loss: 0.0969 - val_mae: 0.0969\n",
      "Epoch 143/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0851 - mae: 0.0851 - val_loss: 0.0757 - val_mae: 0.0757\n",
      "Epoch 144/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0771 - mae: 0.0771 - val_loss: 0.1141 - val_mae: 0.1141\n",
      "Epoch 145/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0788 - mae: 0.0788 - val_loss: 0.1010 - val_mae: 0.1010\n",
      "Epoch 146/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0800 - mae: 0.0800 - val_loss: 0.1098 - val_mae: 0.1098\n",
      "Epoch 147/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0911 - mae: 0.0911 - val_loss: 0.1007 - val_mae: 0.1007\n",
      "Epoch 148/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0833 - mae: 0.0833 - val_loss: 0.0802 - val_mae: 0.0802\n",
      "Epoch 149/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0778 - mae: 0.0778 - val_loss: 0.0918 - val_mae: 0.0918\n",
      "Epoch 150/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0807 - mae: 0.0807 - val_loss: 0.0556 - val_mae: 0.0556\n",
      "Epoch 151/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0655 - mae: 0.0655 - val_loss: 0.0815 - val_mae: 0.0815\n",
      "Epoch 152/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0875 - mae: 0.0875 - val_loss: 0.0962 - val_mae: 0.0962\n",
      "Epoch 153/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0755 - mae: 0.0755 - val_loss: 0.0687 - val_mae: 0.0687\n",
      "Epoch 154/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0786 - mae: 0.0786 - val_loss: 0.0978 - val_mae: 0.0978\n",
      "Epoch 155/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0791 - mae: 0.0791 - val_loss: 0.0604 - val_mae: 0.0604\n",
      "Epoch 156/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0653 - mae: 0.0653 - val_loss: 0.0640 - val_mae: 0.0640\n",
      "Epoch 157/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.0720 - mae: 0.0720 - val_loss: 0.0596 - val_mae: 0.0596\n",
      "Epoch 158/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0824 - mae: 0.0824 - val_loss: 0.0921 - val_mae: 0.0921\n",
      "Epoch 159/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0752 - mae: 0.0752 - val_loss: 0.0816 - val_mae: 0.0816\n",
      "Epoch 160/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0713 - mae: 0.0713 - val_loss: 0.0807 - val_mae: 0.0807\n",
      "Epoch 161/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0653 - mae: 0.0653 - val_loss: 0.0693 - val_mae: 0.0693\n",
      "Epoch 162/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0761 - mae: 0.0761 - val_loss: 0.0847 - val_mae: 0.0847\n",
      "Epoch 163/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0741 - mae: 0.0741 - val_loss: 0.0898 - val_mae: 0.0898\n",
      "Epoch 164/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0747 - mae: 0.0747 - val_loss: 0.0778 - val_mae: 0.0778\n",
      "Epoch 165/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0768 - mae: 0.0768 - val_loss: 0.0664 - val_mae: 0.0664\n",
      "Epoch 166/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0675 - mae: 0.0675 - val_loss: 0.0767 - val_mae: 0.0767\n",
      "Epoch 167/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0763 - mae: 0.0763 - val_loss: 0.0879 - val_mae: 0.0879\n",
      "Epoch 168/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0755 - mae: 0.0755 - val_loss: 0.0650 - val_mae: 0.0650\n",
      "Epoch 169/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0635 - mae: 0.0635 - val_loss: 0.0724 - val_mae: 0.0724\n",
      "Epoch 170/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0674 - mae: 0.0674 - val_loss: 0.0675 - val_mae: 0.0675\n",
      "Epoch 171/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0772 - mae: 0.0772 - val_loss: 0.0866 - val_mae: 0.0866\n",
      "Epoch 172/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0740 - mae: 0.0740 - val_loss: 0.0729 - val_mae: 0.0729\n",
      "Epoch 173/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0676 - mae: 0.0676 - val_loss: 0.0820 - val_mae: 0.0820\n",
      "Epoch 174/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0668 - mae: 0.0668 - val_loss: 0.1102 - val_mae: 0.1102\n",
      "Epoch 175/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0758 - mae: 0.0758 - val_loss: 0.0665 - val_mae: 0.0665\n",
      "Epoch 176/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0709 - mae: 0.0709 - val_loss: 0.0591 - val_mae: 0.0591\n",
      "Epoch 177/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0696 - mae: 0.0696 - val_loss: 0.0745 - val_mae: 0.0745\n",
      "Epoch 178/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0682 - mae: 0.0682 - val_loss: 0.0673 - val_mae: 0.0673\n",
      "Epoch 179/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0740 - mae: 0.0740 - val_loss: 0.0763 - val_mae: 0.0763\n",
      "Epoch 180/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0740 - mae: 0.0740 - val_loss: 0.0743 - val_mae: 0.0743\n",
      "Epoch 181/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0713 - mae: 0.0713 - val_loss: 0.0757 - val_mae: 0.0757\n",
      "Epoch 182/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0733 - mae: 0.0733 - val_loss: 0.0659 - val_mae: 0.0659\n",
      "Epoch 183/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0687 - mae: 0.0687 - val_loss: 0.0600 - val_mae: 0.0600\n",
      "Epoch 184/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0671 - mae: 0.0671 - val_loss: 0.0682 - val_mae: 0.0682\n",
      "Epoch 185/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0646 - mae: 0.0646 - val_loss: 0.0744 - val_mae: 0.0744\n",
      "Epoch 186/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0705 - mae: 0.0705 - val_loss: 0.0625 - val_mae: 0.0625\n",
      "Epoch 187/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0698 - mae: 0.0698 - val_loss: 0.0601 - val_mae: 0.0601\n",
      "Epoch 188/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0672 - mae: 0.0672 - val_loss: 0.0857 - val_mae: 0.0857\n",
      "Epoch 189/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0747 - mae: 0.0747 - val_loss: 0.0707 - val_mae: 0.0707\n",
      "Epoch 190/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0670 - mae: 0.0670 - val_loss: 0.0813 - val_mae: 0.0813\n",
      "Epoch 191/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0690 - mae: 0.0690 - val_loss: 0.0673 - val_mae: 0.0673\n",
      "Epoch 192/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0699 - mae: 0.0699 - val_loss: 0.0774 - val_mae: 0.0774\n",
      "Epoch 193/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0621 - mae: 0.0621 - val_loss: 0.0666 - val_mae: 0.0666\n",
      "Epoch 194/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0731 - mae: 0.0731 - val_loss: 0.0653 - val_mae: 0.0653\n",
      "Epoch 195/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0621 - mae: 0.0621 - val_loss: 0.0868 - val_mae: 0.0868\n",
      "Epoch 196/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0753 - mae: 0.0753 - val_loss: 0.0722 - val_mae: 0.0722\n",
      "Epoch 197/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0631 - mae: 0.0631 - val_loss: 0.0468 - val_mae: 0.0468\n",
      "Epoch 198/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0664 - mae: 0.0664 - val_loss: 0.0742 - val_mae: 0.0742\n",
      "Epoch 199/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0746 - mae: 0.0746 - val_loss: 0.0506 - val_mae: 0.0506\n",
      "Epoch 200/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0742 - mae: 0.0742 - val_loss: 0.0709 - val_mae: 0.0709\n",
      "Epoch 201/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0591 - mae: 0.0591 - val_loss: 0.0646 - val_mae: 0.0646\n",
      "Epoch 202/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0651 - mae: 0.0651 - val_loss: 0.0736 - val_mae: 0.0736\n",
      "Epoch 203/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0750 - mae: 0.0750 - val_loss: 0.0516 - val_mae: 0.0516\n",
      "Epoch 204/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0637 - mae: 0.0637 - val_loss: 0.0806 - val_mae: 0.0806\n",
      "Epoch 205/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0739 - mae: 0.0739 - val_loss: 0.0814 - val_mae: 0.0814\n",
      "Epoch 206/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0679 - mae: 0.0679 - val_loss: 0.0530 - val_mae: 0.0530\n",
      "Epoch 207/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0674 - mae: 0.0674 - val_loss: 0.0753 - val_mae: 0.0753\n",
      "Epoch 208/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0761 - mae: 0.0761 - val_loss: 0.0639 - val_mae: 0.0639\n",
      "Epoch 209/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.0609 - mae: 0.0609 - val_loss: 0.0709 - val_mae: 0.0709\n",
      "Epoch 210/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0691 - mae: 0.0691 - val_loss: 0.0745 - val_mae: 0.0745\n",
      "Epoch 211/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0600 - mae: 0.0600 - val_loss: 0.0673 - val_mae: 0.0673\n",
      "Epoch 212/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0693 - mae: 0.0693 - val_loss: 0.0675 - val_mae: 0.0675\n",
      "Epoch 213/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0608 - mae: 0.0608 - val_loss: 0.0467 - val_mae: 0.0467\n",
      "Epoch 214/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0616 - mae: 0.0616 - val_loss: 0.0786 - val_mae: 0.0786\n",
      "Epoch 215/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0761 - mae: 0.0761 - val_loss: 0.0563 - val_mae: 0.0563\n",
      "Epoch 216/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0670 - mae: 0.0670 - val_loss: 0.0632 - val_mae: 0.0632\n",
      "Epoch 217/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0666 - mae: 0.0666 - val_loss: 0.0658 - val_mae: 0.0658\n",
      "Epoch 218/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0643 - mae: 0.0643 - val_loss: 0.0513 - val_mae: 0.0513\n",
      "Epoch 219/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0652 - mae: 0.0652 - val_loss: 0.0805 - val_mae: 0.0805\n",
      "Epoch 220/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0720 - mae: 0.0720 - val_loss: 0.0687 - val_mae: 0.0687\n",
      "Epoch 221/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0632 - mae: 0.0632 - val_loss: 0.0538 - val_mae: 0.0538\n",
      "Epoch 222/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0536 - mae: 0.0536 - val_loss: 0.0660 - val_mae: 0.0660\n",
      "Epoch 223/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0744 - mae: 0.0744 - val_loss: 0.0683 - val_mae: 0.0683\n",
      "Epoch 224/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0627 - mae: 0.0627 - val_loss: 0.0732 - val_mae: 0.0732\n",
      "Epoch 225/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0763 - mae: 0.0763 - val_loss: 0.0694 - val_mae: 0.0694\n",
      "Epoch 226/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0665 - mae: 0.0665 - val_loss: 0.0510 - val_mae: 0.0510\n",
      "Epoch 227/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0686 - mae: 0.0686 - val_loss: 0.0518 - val_mae: 0.0518\n",
      "Epoch 228/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0581 - mae: 0.0581 - val_loss: 0.0811 - val_mae: 0.0811\n",
      "Epoch 229/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0695 - mae: 0.0695 - val_loss: 0.0623 - val_mae: 0.0623\n",
      "Epoch 230/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0634 - mae: 0.0634 - val_loss: 0.0665 - val_mae: 0.0665\n",
      "Epoch 231/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0681 - mae: 0.0681 - val_loss: 0.0694 - val_mae: 0.0694\n",
      "Epoch 232/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0550 - mae: 0.0550 - val_loss: 0.0615 - val_mae: 0.0615\n",
      "Epoch 233/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0745 - mae: 0.0745 - val_loss: 0.0702 - val_mae: 0.0702\n",
      "Epoch 234/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0710 - mae: 0.0710 - val_loss: 0.0689 - val_mae: 0.0689\n",
      "Epoch 235/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0687 - mae: 0.0687 - val_loss: 0.0579 - val_mae: 0.0579\n",
      "Epoch 236/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0547 - mae: 0.0547 - val_loss: 0.0797 - val_mae: 0.0797\n",
      "Epoch 237/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0738 - mae: 0.0738 - val_loss: 0.0621 - val_mae: 0.0621\n",
      "Epoch 238/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0680 - mae: 0.0680 - val_loss: 0.0671 - val_mae: 0.0671\n",
      "Epoch 239/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0673 - mae: 0.0673 - val_loss: 0.0824 - val_mae: 0.0824\n",
      "Epoch 240/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0668 - mae: 0.0668 - val_loss: 0.0765 - val_mae: 0.0765\n",
      "Epoch 241/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0690 - mae: 0.0690 - val_loss: 0.0563 - val_mae: 0.0563\n",
      "Epoch 242/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0610 - mae: 0.0610 - val_loss: 0.0599 - val_mae: 0.0599\n",
      "Epoch 243/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0687 - mae: 0.0687 - val_loss: 0.0565 - val_mae: 0.0565\n",
      "Epoch 244/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0706 - mae: 0.0706 - val_loss: 0.0622 - val_mae: 0.0622\n",
      "Epoch 245/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0678 - mae: 0.0678 - val_loss: 0.0704 - val_mae: 0.0704\n",
      "Epoch 246/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0644 - mae: 0.0644 - val_loss: 0.0493 - val_mae: 0.0493\n",
      "Epoch 247/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0636 - mae: 0.0636 - val_loss: 0.0753 - val_mae: 0.0753\n",
      "Epoch 248/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0693 - mae: 0.0693 - val_loss: 0.0558 - val_mae: 0.0558\n",
      "Epoch 249/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0483 - mae: 0.0483 - val_loss: 0.0763 - val_mae: 0.0763\n",
      "Epoch 250/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0816 - mae: 0.0816 - val_loss: 0.0782 - val_mae: 0.0782\n",
      "Epoch 251/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.0651 - mae: 0.0651 - val_loss: 0.0635 - val_mae: 0.0635\n",
      "Epoch 252/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0659 - mae: 0.0659 - val_loss: 0.0782 - val_mae: 0.0782\n",
      "Epoch 253/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0666 - mae: 0.0666 - val_loss: 0.0698 - val_mae: 0.0698\n",
      "Epoch 254/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0691 - mae: 0.0691 - val_loss: 0.0625 - val_mae: 0.0625\n",
      "Epoch 255/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0657 - mae: 0.0657 - val_loss: 0.0702 - val_mae: 0.0702\n",
      "Epoch 256/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0580 - mae: 0.0580 - val_loss: 0.0668 - val_mae: 0.0668\n",
      "Epoch 257/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0711 - mae: 0.0711 - val_loss: 0.0629 - val_mae: 0.0629\n",
      "Epoch 258/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0695 - mae: 0.0695 - val_loss: 0.0651 - val_mae: 0.0651\n",
      "Epoch 259/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0666 - mae: 0.0666 - val_loss: 0.0653 - val_mae: 0.0653\n",
      "Epoch 260/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0593 - mae: 0.0593 - val_loss: 0.0604 - val_mae: 0.0604\n",
      "Epoch 261/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0593 - mae: 0.0593 - val_loss: 0.0735 - val_mae: 0.0735\n",
      "Epoch 262/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0746 - mae: 0.0746 - val_loss: 0.0614 - val_mae: 0.0614\n",
      "Epoch 263/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0674 - mae: 0.0674 - val_loss: 0.0755 - val_mae: 0.0755\n",
      "Epoch 264/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0640 - mae: 0.0640 - val_loss: 0.0724 - val_mae: 0.0724\n",
      "Epoch 265/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0644 - mae: 0.0644 - val_loss: 0.0646 - val_mae: 0.0646\n",
      "Epoch 266/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0688 - mae: 0.0688 - val_loss: 0.0667 - val_mae: 0.0667\n",
      "Epoch 267/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0641 - mae: 0.0641 - val_loss: 0.0684 - val_mae: 0.0684\n",
      "Epoch 268/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0612 - mae: 0.0612 - val_loss: 0.0849 - val_mae: 0.0849\n",
      "Epoch 269/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0721 - mae: 0.0721 - val_loss: 0.0649 - val_mae: 0.0649\n",
      "Epoch 270/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0708 - mae: 0.0708 - val_loss: 0.0729 - val_mae: 0.0729\n",
      "Epoch 271/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0655 - mae: 0.0655 - val_loss: 0.0538 - val_mae: 0.0538\n",
      "Epoch 272/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0516 - mae: 0.0516 - val_loss: 0.0900 - val_mae: 0.0900\n",
      "Epoch 273/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0711 - mae: 0.0711 - val_loss: 0.0701 - val_mae: 0.0701\n",
      "Epoch 274/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0617 - mae: 0.0617 - val_loss: 0.0750 - val_mae: 0.0750\n",
      "Epoch 275/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0628 - mae: 0.0628 - val_loss: 0.0565 - val_mae: 0.0565\n",
      "Epoch 276/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0607 - mae: 0.0607 - val_loss: 0.0581 - val_mae: 0.0581\n",
      "Epoch 277/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0689 - mae: 0.0689 - val_loss: 0.0492 - val_mae: 0.0492\n",
      "Epoch 278/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0639 - mae: 0.0639 - val_loss: 0.0656 - val_mae: 0.0656\n",
      "Epoch 279/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0563 - mae: 0.0563 - val_loss: 0.0511 - val_mae: 0.0511\n",
      "Epoch 280/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0630 - mae: 0.0630 - val_loss: 0.0937 - val_mae: 0.0937\n",
      "Epoch 281/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0701 - mae: 0.0701 - val_loss: 0.0556 - val_mae: 0.0556\n",
      "Epoch 282/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0604 - mae: 0.0604 - val_loss: 0.0480 - val_mae: 0.0480\n",
      "Epoch 283/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0566 - mae: 0.0566 - val_loss: 0.0826 - val_mae: 0.0826\n",
      "Epoch 284/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0733 - mae: 0.0733 - val_loss: 0.0623 - val_mae: 0.0623\n",
      "Epoch 285/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0617 - mae: 0.0617 - val_loss: 0.0572 - val_mae: 0.0572\n",
      "Epoch 286/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0494 - mae: 0.0494 - val_loss: 0.0770 - val_mae: 0.0770\n",
      "Epoch 287/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0747 - mae: 0.0747 - val_loss: 0.0744 - val_mae: 0.0744\n",
      "Epoch 288/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0652 - mae: 0.0652 - val_loss: 0.0638 - val_mae: 0.0638\n",
      "Epoch 289/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0688 - mae: 0.0688 - val_loss: 0.0793 - val_mae: 0.0793\n",
      "Epoch 290/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0668 - mae: 0.0668 - val_loss: 0.0679 - val_mae: 0.0679\n",
      "Epoch 291/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0604 - mae: 0.0604 - val_loss: 0.0531 - val_mae: 0.0531\n",
      "Epoch 292/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0649 - mae: 0.0649 - val_loss: 0.0638 - val_mae: 0.0638\n",
      "Epoch 293/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0675 - mae: 0.0675 - val_loss: 0.0717 - val_mae: 0.0717\n",
      "Epoch 294/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0664 - mae: 0.0664 - val_loss: 0.0548 - val_mae: 0.0548\n",
      "Epoch 295/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.0482 - mae: 0.0482 - val_loss: 0.0721 - val_mae: 0.0721\n",
      "Epoch 296/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0717 - mae: 0.0717 - val_loss: 0.0567 - val_mae: 0.0567\n",
      "Epoch 297/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.0635 - mae: 0.0635 - val_loss: 0.0607 - val_mae: 0.0607\n",
      "Epoch 298/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0653 - mae: 0.0653 - val_loss: 0.0627 - val_mae: 0.0627\n",
      "Epoch 299/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0622 - mae: 0.0622 - val_loss: 0.0598 - val_mae: 0.0598\n",
      "Epoch 300/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0635 - mae: 0.0635 - val_loss: 0.0868 - val_mae: 0.0868\n",
      "Epoch 301/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0661 - mae: 0.0661 - val_loss: 0.0573 - val_mae: 0.0573\n",
      "Epoch 302/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0583 - mae: 0.0583 - val_loss: 0.0548 - val_mae: 0.0548\n",
      "Epoch 303/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0693 - mae: 0.0693 - val_loss: 0.0667 - val_mae: 0.0667\n",
      "Epoch 304/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0607 - mae: 0.0607 - val_loss: 0.0770 - val_mae: 0.0770\n",
      "Epoch 305/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0689 - mae: 0.0689 - val_loss: 0.0779 - val_mae: 0.0779\n",
      "Epoch 306/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0651 - mae: 0.0651 - val_loss: 0.0800 - val_mae: 0.0800\n",
      "Epoch 307/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0609 - mae: 0.0609 - val_loss: 0.0756 - val_mae: 0.0756\n",
      "Epoch 308/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0608 - mae: 0.0608 - val_loss: 0.0562 - val_mae: 0.0562\n",
      "Epoch 309/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0726 - mae: 0.0726 - val_loss: 0.0630 - val_mae: 0.0630\n",
      "Epoch 310/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0598 - mae: 0.0598 - val_loss: 0.0570 - val_mae: 0.0570\n",
      "Epoch 311/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0646 - mae: 0.0646 - val_loss: 0.0740 - val_mae: 0.0740\n",
      "Epoch 312/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0634 - mae: 0.0634 - val_loss: 0.0494 - val_mae: 0.0494\n",
      "Epoch 313/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0623 - mae: 0.0623 - val_loss: 0.0692 - val_mae: 0.0692\n",
      "Epoch 314/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.0622 - mae: 0.0622 - val_loss: 0.0584 - val_mae: 0.0584\n",
      "Epoch 315/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0669 - mae: 0.0669 - val_loss: 0.0583 - val_mae: 0.0583\n",
      "Epoch 316/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0636 - mae: 0.0636 - val_loss: 0.0740 - val_mae: 0.0740\n",
      "Epoch 317/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0634 - mae: 0.0634 - val_loss: 0.0626 - val_mae: 0.0626\n",
      "Epoch 318/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0591 - mae: 0.0591 - val_loss: 0.0553 - val_mae: 0.0553\n",
      "Epoch 319/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0655 - mae: 0.0655 - val_loss: 0.0554 - val_mae: 0.0554\n",
      "Epoch 320/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0655 - mae: 0.0655 - val_loss: 0.0590 - val_mae: 0.0590\n",
      "Epoch 321/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0551 - mae: 0.0551 - val_loss: 0.0520 - val_mae: 0.0520\n",
      "Epoch 322/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0604 - mae: 0.0604 - val_loss: 0.0947 - val_mae: 0.0947\n",
      "Epoch 323/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0736 - mae: 0.0736 - val_loss: 0.0739 - val_mae: 0.0739\n",
      "Epoch 324/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0669 - mae: 0.0669 - val_loss: 0.0623 - val_mae: 0.0623\n",
      "Epoch 325/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0578 - mae: 0.0578 - val_loss: 0.0458 - val_mae: 0.0458\n",
      "Epoch 326/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0630 - mae: 0.0630 - val_loss: 0.0764 - val_mae: 0.0764\n",
      "Epoch 327/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0665 - mae: 0.0665 - val_loss: 0.0670 - val_mae: 0.0670\n",
      "Epoch 328/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0654 - mae: 0.0654 - val_loss: 0.0590 - val_mae: 0.0590\n",
      "Epoch 329/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0647 - mae: 0.0647 - val_loss: 0.0642 - val_mae: 0.0642\n",
      "Epoch 330/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0658 - mae: 0.0658 - val_loss: 0.0508 - val_mae: 0.0508\n",
      "Epoch 331/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0591 - mae: 0.0591 - val_loss: 0.0495 - val_mae: 0.0495\n",
      "Epoch 332/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0627 - mae: 0.0627 - val_loss: 0.0610 - val_mae: 0.0610\n",
      "Epoch 333/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0607 - mae: 0.0607 - val_loss: 0.0668 - val_mae: 0.0668\n",
      "Epoch 334/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0652 - mae: 0.0652 - val_loss: 0.0512 - val_mae: 0.0512\n",
      "Epoch 335/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0577 - mae: 0.0577 - val_loss: 0.0713 - val_mae: 0.0713\n",
      "Epoch 336/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0609 - mae: 0.0609 - val_loss: 0.0781 - val_mae: 0.0781\n",
      "Epoch 337/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0658 - mae: 0.0658 - val_loss: 0.0684 - val_mae: 0.0684\n",
      "Epoch 338/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0638 - mae: 0.0638 - val_loss: 0.0623 - val_mae: 0.0623\n",
      "Epoch 339/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0587 - mae: 0.0587 - val_loss: 0.0551 - val_mae: 0.0551\n",
      "Epoch 340/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0615 - mae: 0.0615 - val_loss: 0.0595 - val_mae: 0.0595\n",
      "Epoch 341/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0661 - mae: 0.0661 - val_loss: 0.0447 - val_mae: 0.0447\n",
      "Epoch 342/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0512 - mae: 0.0512 - val_loss: 0.0568 - val_mae: 0.0568\n",
      "Epoch 343/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0635 - mae: 0.0635 - val_loss: 0.0717 - val_mae: 0.0717\n",
      "Epoch 344/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0694 - mae: 0.0694 - val_loss: 0.0492 - val_mae: 0.0492\n",
      "Epoch 345/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0575 - mae: 0.0575 - val_loss: 0.0681 - val_mae: 0.0681\n",
      "Epoch 346/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0701 - mae: 0.0701 - val_loss: 0.0573 - val_mae: 0.0573\n",
      "Epoch 347/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0522 - mae: 0.0522 - val_loss: 0.0467 - val_mae: 0.0467\n",
      "Epoch 348/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0595 - mae: 0.0595 - val_loss: 0.0671 - val_mae: 0.0671\n",
      "Epoch 349/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0676 - mae: 0.0676 - val_loss: 0.0491 - val_mae: 0.0491\n",
      "Epoch 350/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0592 - mae: 0.0592 - val_loss: 0.0553 - val_mae: 0.0553\n",
      "Epoch 351/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0596 - mae: 0.0596 - val_loss: 0.0876 - val_mae: 0.0876\n",
      "Epoch 352/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0671 - mae: 0.0671 - val_loss: 0.0548 - val_mae: 0.0548\n",
      "Epoch 353/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0529 - mae: 0.0529 - val_loss: 0.0753 - val_mae: 0.0753\n",
      "Epoch 354/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0689 - mae: 0.0689 - val_loss: 0.0817 - val_mae: 0.0817\n",
      "Epoch 355/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0614 - mae: 0.0614 - val_loss: 0.0557 - val_mae: 0.0557\n",
      "Epoch 356/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0576 - mae: 0.0576 - val_loss: 0.0642 - val_mae: 0.0642\n",
      "Epoch 357/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0652 - mae: 0.0652 - val_loss: 0.0677 - val_mae: 0.0677\n",
      "Epoch 358/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0738 - mae: 0.0738 - val_loss: 0.0530 - val_mae: 0.0530\n",
      "Epoch 359/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0575 - mae: 0.0575 - val_loss: 0.0557 - val_mae: 0.0557\n",
      "Epoch 360/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0586 - mae: 0.0586 - val_loss: 0.0590 - val_mae: 0.0590\n",
      "Epoch 361/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0531 - mae: 0.0531 - val_loss: 0.0709 - val_mae: 0.0709\n",
      "Epoch 362/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0703 - mae: 0.0703 - val_loss: 0.0745 - val_mae: 0.0745\n",
      "Epoch 363/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0718 - mae: 0.0718 - val_loss: 0.0492 - val_mae: 0.0492\n",
      "Epoch 364/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0606 - mae: 0.0606 - val_loss: 0.0701 - val_mae: 0.0701\n",
      "Epoch 365/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0637 - mae: 0.0637 - val_loss: 0.0575 - val_mae: 0.0575\n",
      "Epoch 366/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0563 - mae: 0.0563 - val_loss: 0.0575 - val_mae: 0.0575\n",
      "Epoch 367/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0568 - mae: 0.0568 - val_loss: 0.0553 - val_mae: 0.0553\n",
      "Epoch 368/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0616 - mae: 0.0616 - val_loss: 0.0565 - val_mae: 0.0565\n",
      "Epoch 369/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0635 - mae: 0.0635 - val_loss: 0.0578 - val_mae: 0.0578\n",
      "Epoch 370/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0587 - mae: 0.0587 - val_loss: 0.0581 - val_mae: 0.0581\n",
      "Epoch 371/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0615 - mae: 0.0615 - val_loss: 0.0678 - val_mae: 0.0678\n",
      "Epoch 372/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0616 - mae: 0.0616 - val_loss: 0.0546 - val_mae: 0.0546\n",
      "Epoch 373/500\n",
      "6/6 [==============================] - 0s 8ms/step - loss: 0.0511 - mae: 0.0511 - val_loss: 0.0865 - val_mae: 0.0865\n",
      "Epoch 374/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0692 - mae: 0.0692 - val_loss: 0.0768 - val_mae: 0.0768\n",
      "Epoch 375/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0722 - mae: 0.0722 - val_loss: 0.0623 - val_mae: 0.0623\n",
      "Epoch 376/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0618 - mae: 0.0618 - val_loss: 0.0610 - val_mae: 0.0610\n",
      "Epoch 377/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0577 - mae: 0.0577 - val_loss: 0.0709 - val_mae: 0.0709\n",
      "Epoch 378/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0587 - mae: 0.0587 - val_loss: 0.0464 - val_mae: 0.0464\n",
      "Epoch 379/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0532 - mae: 0.0532 - val_loss: 0.0588 - val_mae: 0.0588\n",
      "Epoch 380/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0666 - mae: 0.0666 - val_loss: 0.0670 - val_mae: 0.0670\n",
      "Epoch 381/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0672 - mae: 0.0672 - val_loss: 0.0543 - val_mae: 0.0543\n",
      "Epoch 382/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0547 - mae: 0.0547 - val_loss: 0.0766 - val_mae: 0.0766\n",
      "Epoch 383/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0630 - mae: 0.0630 - val_loss: 0.0728 - val_mae: 0.0728\n",
      "Epoch 384/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0563 - mae: 0.0563 - val_loss: 0.0604 - val_mae: 0.0604\n",
      "Epoch 385/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0638 - mae: 0.0638 - val_loss: 0.0703 - val_mae: 0.0703\n",
      "Epoch 386/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0672 - mae: 0.0672 - val_loss: 0.0755 - val_mae: 0.0755\n",
      "Epoch 387/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0637 - mae: 0.0637 - val_loss: 0.0572 - val_mae: 0.0572\n",
      "Epoch 388/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0595 - mae: 0.0595 - val_loss: 0.0591 - val_mae: 0.0591\n",
      "Epoch 389/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0606 - mae: 0.0606 - val_loss: 0.0664 - val_mae: 0.0664\n",
      "Epoch 390/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0662 - mae: 0.0662 - val_loss: 0.0504 - val_mae: 0.0504\n",
      "Epoch 391/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0582 - mae: 0.0582 - val_loss: 0.0628 - val_mae: 0.0628\n",
      "Epoch 392/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0619 - mae: 0.0619 - val_loss: 0.0706 - val_mae: 0.0706\n",
      "Epoch 393/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0631 - mae: 0.0631 - val_loss: 0.0687 - val_mae: 0.0687\n",
      "Epoch 394/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0592 - mae: 0.0592 - val_loss: 0.0576 - val_mae: 0.0576\n",
      "Epoch 395/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0564 - mae: 0.0564 - val_loss: 0.0719 - val_mae: 0.0719\n",
      "Epoch 396/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0611 - mae: 0.0611 - val_loss: 0.0549 - val_mae: 0.0549\n",
      "Epoch 397/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0615 - mae: 0.0615 - val_loss: 0.0558 - val_mae: 0.0558\n",
      "Epoch 398/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0631 - mae: 0.0631 - val_loss: 0.0566 - val_mae: 0.0566\n",
      "Epoch 399/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0625 - mae: 0.0625 - val_loss: 0.0757 - val_mae: 0.0757\n",
      "Epoch 400/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0633 - mae: 0.0633 - val_loss: 0.0545 - val_mae: 0.0545\n",
      "Epoch 401/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0581 - mae: 0.0581 - val_loss: 0.0530 - val_mae: 0.0530\n",
      "Epoch 402/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0590 - mae: 0.0590 - val_loss: 0.0672 - val_mae: 0.0672\n",
      "Epoch 403/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0580 - mae: 0.0580 - val_loss: 0.0481 - val_mae: 0.0481\n",
      "Epoch 404/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0590 - mae: 0.0590 - val_loss: 0.0713 - val_mae: 0.0713\n",
      "Epoch 405/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0594 - mae: 0.0594 - val_loss: 0.0767 - val_mae: 0.0767\n",
      "Epoch 406/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0658 - mae: 0.0658 - val_loss: 0.0601 - val_mae: 0.0601\n",
      "Epoch 407/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.0566 - mae: 0.0566 - val_loss: 0.0638 - val_mae: 0.0638\n",
      "Epoch 408/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0565 - mae: 0.0565 - val_loss: 0.0510 - val_mae: 0.0510\n",
      "Epoch 409/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0610 - mae: 0.0610 - val_loss: 0.0487 - val_mae: 0.0487\n",
      "Epoch 410/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0597 - mae: 0.0597 - val_loss: 0.0550 - val_mae: 0.0550\n",
      "Epoch 411/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0479 - mae: 0.0479 - val_loss: 0.0783 - val_mae: 0.0783\n",
      "Epoch 412/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0689 - mae: 0.0689 - val_loss: 0.0786 - val_mae: 0.0786\n",
      "Epoch 413/500\n",
      "6/6 [==============================] - 0s 7ms/step - loss: 0.0648 - mae: 0.0648 - val_loss: 0.0482 - val_mae: 0.0482\n",
      "Epoch 414/500\n",
      "6/6 [==============================] - 0s 8ms/step - loss: 0.0483 - mae: 0.0483 - val_loss: 0.0504 - val_mae: 0.0504\n",
      "Epoch 415/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0649 - mae: 0.0649 - val_loss: 0.0641 - val_mae: 0.0641\n",
      "Epoch 416/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0596 - mae: 0.0596 - val_loss: 0.0569 - val_mae: 0.0569\n",
      "Epoch 417/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0624 - mae: 0.0624 - val_loss: 0.0635 - val_mae: 0.0635\n",
      "Epoch 418/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0621 - mae: 0.0621 - val_loss: 0.0504 - val_mae: 0.0504\n",
      "Epoch 419/500\n",
      "6/6 [==============================] - 0s 4ms/step - loss: 0.0575 - mae: 0.0575 - val_loss: 0.0558 - val_mae: 0.0558\n",
      "Epoch 420/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0565 - mae: 0.0565 - val_loss: 0.0818 - val_mae: 0.0818\n",
      "Epoch 421/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0649 - mae: 0.0649 - val_loss: 0.0672 - val_mae: 0.0672\n",
      "Epoch 422/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0567 - mae: 0.0567 - val_loss: 0.0448 - val_mae: 0.0448\n",
      "Epoch 423/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0549 - mae: 0.0549 - val_loss: 0.0880 - val_mae: 0.0880\n",
      "Epoch 424/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0631 - mae: 0.0631 - val_loss: 0.0589 - val_mae: 0.0589\n",
      "Epoch 425/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0619 - mae: 0.0619 - val_loss: 0.0746 - val_mae: 0.0746\n",
      "Epoch 426/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0622 - mae: 0.0622 - val_loss: 0.0582 - val_mae: 0.0582\n",
      "Epoch 427/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0568 - mae: 0.0568 - val_loss: 0.0626 - val_mae: 0.0626\n",
      "Epoch 428/500\n",
      "6/6 [==============================] - 0s 4ms/step - loss: 0.0586 - mae: 0.0586 - val_loss: 0.0560 - val_mae: 0.0560\n",
      "Epoch 429/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0638 - mae: 0.0638 - val_loss: 0.0606 - val_mae: 0.0606\n",
      "Epoch 430/500\n",
      "6/6 [==============================] - 0s 4ms/step - loss: 0.0630 - mae: 0.0630 - val_loss: 0.0694 - val_mae: 0.0694\n",
      "Epoch 431/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0508 - mae: 0.0508 - val_loss: 0.0582 - val_mae: 0.0582\n",
      "Epoch 432/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0648 - mae: 0.0648 - val_loss: 0.0693 - val_mae: 0.0693\n",
      "Epoch 433/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0633 - mae: 0.0633 - val_loss: 0.0548 - val_mae: 0.0548\n",
      "Epoch 434/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0560 - mae: 0.0560 - val_loss: 0.0480 - val_mae: 0.0480\n",
      "Epoch 435/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0678 - mae: 0.0678 - val_loss: 0.0547 - val_mae: 0.0547\n",
      "Epoch 436/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0634 - mae: 0.0634 - val_loss: 0.0652 - val_mae: 0.0652\n",
      "Epoch 437/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0619 - mae: 0.0619 - val_loss: 0.0605 - val_mae: 0.0605\n",
      "Epoch 438/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0593 - mae: 0.0593 - val_loss: 0.0535 - val_mae: 0.0535\n",
      "Epoch 439/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0518 - mae: 0.0518 - val_loss: 0.0491 - val_mae: 0.0491\n",
      "Epoch 440/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0614 - mae: 0.0614 - val_loss: 0.0753 - val_mae: 0.0753\n",
      "Epoch 441/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0638 - mae: 0.0638 - val_loss: 0.0600 - val_mae: 0.0600\n",
      "Epoch 442/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0551 - mae: 0.0551 - val_loss: 0.0870 - val_mae: 0.0870\n",
      "Epoch 443/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0617 - mae: 0.0617 - val_loss: 0.0814 - val_mae: 0.0814\n",
      "Epoch 444/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0678 - mae: 0.0678 - val_loss: 0.0544 - val_mae: 0.0544\n",
      "Epoch 445/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0589 - mae: 0.0589 - val_loss: 0.0709 - val_mae: 0.0709\n",
      "Epoch 446/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0553 - mae: 0.0553 - val_loss: 0.0498 - val_mae: 0.0498\n",
      "Epoch 447/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0580 - mae: 0.0580 - val_loss: 0.0719 - val_mae: 0.0719\n",
      "Epoch 448/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0616 - mae: 0.0616 - val_loss: 0.0577 - val_mae: 0.0577\n",
      "Epoch 449/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0581 - mae: 0.0581 - val_loss: 0.0520 - val_mae: 0.0520\n",
      "Epoch 450/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0515 - mae: 0.0515 - val_loss: 0.0510 - val_mae: 0.0510\n",
      "Epoch 451/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0628 - mae: 0.0628 - val_loss: 0.0813 - val_mae: 0.0813\n",
      "Epoch 452/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0645 - mae: 0.0645 - val_loss: 0.0619 - val_mae: 0.0619\n",
      "Epoch 453/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0546 - mae: 0.0546 - val_loss: 0.0633 - val_mae: 0.0633\n",
      "Epoch 454/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0583 - mae: 0.0583 - val_loss: 0.0859 - val_mae: 0.0859\n",
      "Epoch 455/500\n",
      "6/6 [==============================] - 0s 8ms/step - loss: 0.0652 - mae: 0.0652 - val_loss: 0.0601 - val_mae: 0.0601\n",
      "Epoch 456/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0532 - mae: 0.0532 - val_loss: 0.0462 - val_mae: 0.0462\n",
      "Epoch 457/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0595 - mae: 0.0595 - val_loss: 0.0604 - val_mae: 0.0604\n",
      "Epoch 458/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0532 - mae: 0.0532 - val_loss: 0.0690 - val_mae: 0.0690\n",
      "Epoch 459/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0653 - mae: 0.0653 - val_loss: 0.0606 - val_mae: 0.0606\n",
      "Epoch 460/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0579 - mae: 0.0579 - val_loss: 0.0475 - val_mae: 0.0475\n",
      "Epoch 461/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0518 - mae: 0.0518 - val_loss: 0.0660 - val_mae: 0.0660\n",
      "Epoch 462/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0693 - mae: 0.0693 - val_loss: 0.0497 - val_mae: 0.0497\n",
      "Epoch 463/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0598 - mae: 0.0598 - val_loss: 0.0580 - val_mae: 0.0580\n",
      "Epoch 464/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0596 - mae: 0.0596 - val_loss: 0.0576 - val_mae: 0.0576\n",
      "Epoch 465/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0557 - mae: 0.0557 - val_loss: 0.0559 - val_mae: 0.0559\n",
      "Epoch 466/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0587 - mae: 0.0587 - val_loss: 0.0565 - val_mae: 0.0565\n",
      "Epoch 467/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0575 - mae: 0.0575 - val_loss: 0.0514 - val_mae: 0.0514\n",
      "Epoch 468/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0646 - mae: 0.0646 - val_loss: 0.0438 - val_mae: 0.0438\n",
      "Epoch 469/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0509 - mae: 0.0509 - val_loss: 0.0696 - val_mae: 0.0696\n",
      "Epoch 470/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0589 - mae: 0.0589 - val_loss: 0.0597 - val_mae: 0.0597\n",
      "Epoch 471/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0675 - mae: 0.0675 - val_loss: 0.0624 - val_mae: 0.0624\n",
      "Epoch 472/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0573 - mae: 0.0573 - val_loss: 0.0456 - val_mae: 0.0456\n",
      "Epoch 473/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0509 - mae: 0.0509 - val_loss: 0.0446 - val_mae: 0.0446\n",
      "Epoch 474/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0582 - mae: 0.0582 - val_loss: 0.0651 - val_mae: 0.0651\n",
      "Epoch 475/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0609 - mae: 0.0609 - val_loss: 0.0621 - val_mae: 0.0621\n",
      "Epoch 476/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0569 - mae: 0.0569 - val_loss: 0.0535 - val_mae: 0.0535\n",
      "Epoch 477/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0531 - mae: 0.0531 - val_loss: 0.0915 - val_mae: 0.0915\n",
      "Epoch 478/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0618 - mae: 0.0618 - val_loss: 0.0628 - val_mae: 0.0628\n",
      "Epoch 479/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0596 - mae: 0.0596 - val_loss: 0.0879 - val_mae: 0.0879\n",
      "Epoch 480/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0691 - mae: 0.0691 - val_loss: 0.0614 - val_mae: 0.0614\n",
      "Epoch 481/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0566 - mae: 0.0566 - val_loss: 0.0545 - val_mae: 0.0545\n",
      "Epoch 482/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0478 - mae: 0.0478 - val_loss: 0.0759 - val_mae: 0.0759\n",
      "Epoch 483/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0605 - mae: 0.0605 - val_loss: 0.0812 - val_mae: 0.0812\n",
      "Epoch 484/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0569 - mae: 0.0569 - val_loss: 0.0676 - val_mae: 0.0676\n",
      "Epoch 485/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0645 - mae: 0.0645 - val_loss: 0.0610 - val_mae: 0.0610\n",
      "Epoch 486/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0628 - mae: 0.0628 - val_loss: 0.0500 - val_mae: 0.0500\n",
      "Epoch 487/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0554 - mae: 0.0554 - val_loss: 0.0549 - val_mae: 0.0549\n",
      "Epoch 488/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0570 - mae: 0.0570 - val_loss: 0.0629 - val_mae: 0.0629\n",
      "Epoch 489/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0638 - mae: 0.0638 - val_loss: 0.0601 - val_mae: 0.0601\n",
      "Epoch 490/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0593 - mae: 0.0593 - val_loss: 0.0557 - val_mae: 0.0557\n",
      "Epoch 491/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0616 - mae: 0.0616 - val_loss: 0.0448 - val_mae: 0.0448\n",
      "Epoch 492/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0505 - mae: 0.0505 - val_loss: 0.0642 - val_mae: 0.0642\n",
      "Epoch 493/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0596 - mae: 0.0596 - val_loss: 0.0461 - val_mae: 0.0461\n",
      "Epoch 494/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0580 - mae: 0.0580 - val_loss: 0.0480 - val_mae: 0.0480\n",
      "Epoch 495/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0491 - mae: 0.0491 - val_loss: 0.0800 - val_mae: 0.0800\n",
      "Epoch 496/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0705 - mae: 0.0705 - val_loss: 0.0491 - val_mae: 0.0491\n",
      "Epoch 497/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0589 - mae: 0.0589 - val_loss: 0.0471 - val_mae: 0.0471\n",
      "Epoch 498/500\n",
      "6/6 [==============================] - 0s 6ms/step - loss: 0.0530 - mae: 0.0530 - val_loss: 0.0592 - val_mae: 0.0592\n",
      "Epoch 499/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0579 - mae: 0.0579 - val_loss: 0.0552 - val_mae: 0.0552\n",
      "Epoch 500/500\n",
      "6/6 [==============================] - 0s 5ms/step - loss: 0.0553 - mae: 0.0553 - val_loss: 0.0801 - val_mae: 0.0801\n"
     ]
    }
   ],
   "source": [
    "with tf.device('/GPU:0'):\n",
    "  history = model.fit(x_train,\n",
    "                    y_train,\n",
    "                    epochs=500,\n",
    "                    batch_size=100,\n",
    "                    validation_data=(x_val, y_val),\n",
    "                  )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "TTabUvyrBr3G"
   },
   "source": [
    "**Observe training history**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 452
    },
    "executionInfo": {
     "elapsed": 1926,
     "status": "ok",
     "timestamp": 1691563359874,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "I_cByp7fBBpy",
    "outputId": "aaed40cb-3f4c-4762-b61b-70c79cfcf775"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "loss = history.history['loss']\n",
    "val_loss = history.history['val_loss']\n",
    "\n",
    "epochs = range(1, len(loss) + 1)\n",
    "\n",
    "plt.plot(epochs, loss, 'g', label='Training loss')\n",
    "plt.plot(epochs, val_loss, 'b', label='Validation loss')\n",
    "plt.title('Training and validation loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_LVYt4PSCS_K"
   },
   "source": [
    "**plot the predictions against actual values**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 469
    },
    "executionInfo": {
     "elapsed": 1092,
     "status": "ok",
     "timestamp": 1691563540943,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "-4fln_g2CPER",
    "outputId": "2c843a78-456f-4695-b0ca-26090a2947db"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7/7 [==============================] - 0s 2ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "predictions = model.predict(x_test)\n",
    "\n",
    "plt.clf()\n",
    "plt.title('Comparison of predictions to actual values')\n",
    "plt.plot(x_test, y_test, 'g.', label='Actual')\n",
    "plt.plot(x_test, predictions, 'b.', label='Prediction')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LSviIoLeC_nc"
   },
   "source": [
    "**Convert and optimize it as a tflite model**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 3139,
     "status": "ok",
     "timestamp": 1691563794373,
     "user": {
      "displayName": "Sravan 21BLC1018",
      "userId": "15786248523050194906"
     },
     "user_tz": -330
    },
    "id": "88cag9rqDGoZ",
    "outputId": "60db9c1b-e8b1-464a-a20d-9ca1f872fa56"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /tmp/tmpb8zdobza/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: /tmp/tmpb8zdobza/assets\n",
      "2023-12-18 15:49:55.661558: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:378] Ignored output_format.\n",
      "2023-12-18 15:49:55.661580: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:381] Ignored drop_control_dependency.\n",
      "2023-12-18 15:49:55.661691: I tensorflow/cc/saved_model/reader.cc:83] Reading SavedModel from: /tmp/tmpb8zdobza\n",
      "2023-12-18 15:49:55.662428: I tensorflow/cc/saved_model/reader.cc:51] Reading meta graph with tags { serve }\n",
      "2023-12-18 15:49:55.662438: I tensorflow/cc/saved_model/reader.cc:146] Reading SavedModel debug info (if present) from: /tmp/tmpb8zdobza\n",
      "2023-12-18 15:49:55.664560: I tensorflow/cc/saved_model/loader.cc:233] Restoring SavedModel bundle.\n",
      "2023-12-18 15:49:55.692476: I tensorflow/cc/saved_model/loader.cc:217] Running initialization op on SavedModel bundle at path: /tmp/tmpb8zdobza\n",
      "2023-12-18 15:49:55.700617: I tensorflow/cc/saved_model/loader.cc:316] SavedModel load for tags { serve }; Status: success: OK. Took 38926 microseconds.\n",
      "Summary on the non-converted ops:\n",
      "---------------------------------\n",
      " * Accepted dialects: tfl, builtin, func\n",
      " * Non-Converted Ops: 9, Total Ops 18, % non-converted = 50.00 %\n",
      " * 9 ARITH ops\n",
      "\n",
      "- arith.constant:    9 occurrences  (f32: 9)\n",
      "\n",
      "\n",
      "\n",
      "  (f32: 5)\n",
      "  (uq_8: 1)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "8416"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "converter = tf.lite.TFLiteConverter.from_keras_model(model)\n",
    "converter.optimizations = [tf.lite.Optimize.DEFAULT]\n",
    "tflite_model = converter.convert()\n",
    "\n",
    "open(tflite_model_name + '.tflite', 'wb').write(tflite_model)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "authorship_tag": "ABX9TyP9VuAi2bgXmTvRr9k8U4l8",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}