SVC_Neural_Network.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from urllib.request import Request, urlopen\n",
    "from bs4 import BeautifulSoup\n",
    "import seaborn as sns\n",
    "import re\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from tensorflow.python.keras.models import Sequential\n",
    "from tensorflow.python.keras.layers import Dense, Activation, Dropout\n",
    "from tensorflow.python.keras.wrappers.scikit_learn import KerasRegressor\n",
    "from tensorflow.python.keras.activations import relu, sigmoid, linear, exponential, softmax, tanh\n",
    "\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.exceptions import FitFailedWarning\n",
    "import warnings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def loadData():\n",
    "    data = pd.read_csv('https://www.dropbox.com/s/xn60mr1v40g09hb/major_data_normalized.csv?dl=1')\n",
    "    x = (data[['2','3','4']])\n",
    "    x = x.to_numpy()\n",
    "    x = x.astype(float)\n",
    "    y = (data[['0']])\n",
    "    y = y.to_numpy()\n",
    "    y = y.astype(float)\n",
    "    y = np.reshape(y, (-1,1))\n",
    "\n",
    "    return x,y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plotData(x, y, xlabel, ylabel):\n",
    "    plt.xlabel(xlabel)\n",
    "    plt.ylabel(ylabel)\n",
    "    plt.scatter([x], [y], marker = 'o')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MinMaxScaler(copy=True, feature_range=(0, 1))\n",
      "MinMaxScaler(copy=True, feature_range=(0, 1))\n"
     ]
    }
   ],
   "source": [
    "x, y = loadData()\n",
    "\n",
    "scaler_x = MinMaxScaler()\n",
    "scaler_y = MinMaxScaler()\n",
    "print(scaler_x.fit(x))\n",
    "xscale=scaler_x.transform(x)\n",
    "print(scaler_y.fit(y))\n",
    "yscale=scaler_y.transform(y)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(xscale, yscale)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.wrappers.scikit_learn.KerasRegressor at 0x1a4bf11e90>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def create_model(layers, activation):\n",
    "    model = Sequential()\n",
    "    for i, nodes in enumerate(layers):\n",
    "   \n",
    "        if(i == 0):\n",
    "            model.add(Dense(nodes, input_dim = 3, kernel_initializer = 'normal', activation=activation))\n",
    "            model.add(Dropout(0.25))\n",
    "        else:\n",
    "            model.add(Dense(nodes, activation = activation))\n",
    "      \n",
    "\n",
    "    model.add(Dense(1, activation = 'linear'))\n",
    "\n",
    "    model.compile(loss = 'mse', optimizer = 'adam', metrics = ['mse', 'mae'])\n",
    "\n",
    "    return model\n",
    "\n",
    "model_checked = KerasRegressor(build_fn = create_model, verbose = 0)\n",
    "model_checked"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import svm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(64, input_dim = 3, kernel_initializer = 'normal', activation = 'sigmoid'))\n",
    "model.add(Dropout(0.25))\n",
    "model.add(Dense(32, activation = 'sigmoid'))\n",
    "model.add(Dense(16, activation  = 'sigmoid'))\n",
    "model.add(Dense(1, activation = 'linear'))\n",
    "model.compile(loss = 'mse', optimizer = 'adam', metrics = ['mse', 'mae'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 12 samples, validate on 4 samples\n",
      "Epoch 1/10\n",
      "12/12 [==============================] - 1s 118ms/sample - loss: 0.1166 - mse: 0.1166 - mae: 0.3084 - val_loss: 0.1401 - val_mse: 0.1401 - val_mae: 0.3551\n",
      "Epoch 2/10\n",
      "12/12 [==============================] - 0s 6ms/sample - loss: 0.1008 - mse: 0.1008 - mae: 0.2887 - val_loss: 0.1151 - val_mse: 0.1151 - val_mae: 0.3179\n",
      "Epoch 3/10\n",
      "12/12 [==============================] - 0s 6ms/sample - loss: 0.0953 - mse: 0.0953 - mae: 0.2728 - val_loss: 0.0933 - val_mse: 0.0933 - val_mae: 0.2815\n",
      "Epoch 4/10\n",
      "12/12 [==============================] - 0s 6ms/sample - loss: 0.0829 - mse: 0.0829 - mae: 0.2531 - val_loss: 0.0749 - val_mse: 0.0749 - val_mae: 0.2468\n",
      "Epoch 5/10\n",
      "12/12 [==============================] - 0s 6ms/sample - loss: 0.0809 - mse: 0.0809 - mae: 0.2371 - val_loss: 0.0599 - val_mse: 0.0599 - val_mae: 0.2142\n",
      "Epoch 6/10\n",
      "12/12 [==============================] - 0s 6ms/sample - loss: 0.0716 - mse: 0.0716 - mae: 0.2129 - val_loss: 0.0480 - val_mse: 0.0480 - val_mae: 0.1901\n",
      "Epoch 7/10\n",
      "12/12 [==============================] - 0s 7ms/sample - loss: 0.0694 - mse: 0.0694 - mae: 0.1987 - val_loss: 0.0389 - val_mse: 0.0389 - val_mae: 0.1769\n",
      "Epoch 8/10\n",
      "12/12 [==============================] - 0s 7ms/sample - loss: 0.0734 - mse: 0.0734 - mae: 0.1959 - val_loss: 0.0325 - val_mse: 0.0325 - val_mae: 0.1660\n",
      "Epoch 9/10\n",
      "12/12 [==============================] - 0s 7ms/sample - loss: 0.0697 - mse: 0.0697 - mae: 0.1924 - val_loss: 0.0280 - val_mse: 0.0280 - val_mae: 0.1570\n",
      "Epoch 10/10\n",
      "12/12 [==============================] - 0s 8ms/sample - loss: 0.0752 - mse: 0.0752 - mae: 0.1854 - val_loss: 0.0251 - val_mse: 0.0251 - val_mae: 0.1505\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train, y_train, verbose = 1, validation_split = 0.2, batch_size = 52, epochs = 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['loss', 'mse', 'mae', 'val_loss', 'val_mse', 'val_mae'])\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(history.history.keys())\n",
    "#Loss\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc = 'upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x, y = loadData()\n",
    "plotData(x[:, 0], y, 'Jobs per 1000', 'Total Number of Employees')\n",
    "plt.figure()\n",
    "plotData(x[:, 1], y, 'Location Variable', 'Total Number of Employees')\n",
    "plt.figure()\n",
    "plotData(x[:, 2], y, 'Hourly Wage Average', 'Total Number of Employees')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 [0.2931022]\n",
      "2 [0.293144]\n",
      "3 [0.29285792]\n",
      "4 [0.29294178]\n",
      "5 [0.29294148]\n",
      "6 [0.29346836]\n",
      "7 [0.2929357]\n",
      "8 [0.29361886]\n",
      "9 [0.29331505]\n",
      "10 [0.29339847]\n",
      "11 [0.29368895]\n",
      "12 [0.293527]\n",
      "13 [0.29380894]\n",
      "14 [0.29361078]\n",
      "15 [0.29369986]\n",
      "16 [0.29365772]\n",
      "17 [0.29393405]\n",
      "18 [0.29370475]\n",
      "19 [0.29333776]\n",
      "20 [0.29352126]\n",
      "21 [0.2938934]\n",
      "22 [0.29378468]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x, y = loadData()\n",
    "x = np.expand_dims(x, axis = 1)\n",
    "predictions = []\n",
    "for i in range(len(x)):\n",
    "    x[i]= scaler_x.transform(x[i].reshape(1,-1))\n",
    "    ynew= model.predict(x[i])\n",
    "    #invert normalize\n",
    "    ynew = scaler_y.inverse_transform(ynew.reshape(1,-1))\n",
    "    x[i] = scaler_x.inverse_transform(x[i])\n",
    "    predictions.append(ynew[0])\n",
    "##print(\"X=%s, Predicted=%s\" % (Xnew[0], ynew[0]))\n",
    "plt.figure()\n",
    "plotData(y, predictions, 'measured', 'predicted' )\n",
    "for i in range(len(predictions)):\n",
    "    print(i+1, predictions[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdin",
     "output_type": "stream",
     "text": [
      "Enter the jobs per 1000 people normalized:  .4\n",
      "Enter the location quotient normalized:  .9\n",
      "Enter the average hourly wage normalzied:  .6\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X:  [0.4 0.9 0.6] Predicted =  [0.29300436]\n"
     ]
    }
   ],
   "source": [
    "pop = float(input(\"Enter the jobs per 1000 people normalized: \"))\n",
    "dense = float(input(\"Enter the location quotient normalized: \"))\n",
    "days = float(input(\"Enter the average hourly wage normalzied: \"))\n",
    "x_spec = [[pop, dense, days]]\n",
    "x_spec = scaler_x.transform(x_spec)\n",
    "y_hat = model.predict(x_spec)\n",
    "y_hat = scaler_y.inverse_transform(y_hat.reshape(-1,1))\n",
    "x_spec = scaler_x.inverse_transform(x_spec)\n",
    "print('X: ',x_spec[0], 'Predicted = ', y_hat[0])    #Is there a way to un normalize data quickly?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Non normalized input data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "def loadData1():\n",
    "    data = pd.read_csv('https://www.dropbox.com/s/fotppg0aakiwa0v/major_groups_unnormalized.csv?dl=1')\n",
    "    x = (data[['JOBS_1000','LOC_Q','H_MEAN']])\n",
    "    x = x.to_numpy()\n",
    "    x = x.astype(float)\n",
    "    y = (data[['TOT_EMP']])\n",
    "    y = y.to_numpy()\n",
    "    y = y.astype(float)\n",
    "    y = np.reshape(y, (-1,1))\n",
    "\n",
    "    return x,y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MinMaxScaler(copy=True, feature_range=(0, 1))\n",
      "MinMaxScaler(copy=True, feature_range=(0, 1))\n"
     ]
    }
   ],
   "source": [
    "x, y = loadData1()\n",
    "\n",
    "scaler_x = MinMaxScaler()\n",
    "scaler_y = MinMaxScaler()\n",
    "print(scaler_x.fit(x))\n",
    "xscale=scaler_x.transform(x)\n",
    "print(scaler_y.fit(y))\n",
    "yscale=scaler_y.transform(y)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(xscale, yscale)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(64, input_dim = 3, kernel_initializer = 'normal', activation = 'sigmoid'))\n",
    "model.add(Dropout(0.25))\n",
    "model.add(Dense(32, activation = 'sigmoid'))\n",
    "model.add(Dense(16, activation  = 'sigmoid'))\n",
    "model.add(Dense(1, activation = 'linear'))\n",
    "model.compile(loss = 'mse', optimizer = 'adam', metrics = ['mse', 'mae'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 12 samples, validate on 4 samples\n",
      "Epoch 1/10\n",
      "12/12 [==============================] - 1s 118ms/sample - loss: 1.3904 - mse: 1.3904 - mae: 1.1481 - val_loss: 1.7199 - val_mse: 1.7199 - val_mae: 1.2894\n",
      "Epoch 2/10\n",
      "12/12 [==============================] - 0s 6ms/sample - loss: 1.2415 - mse: 1.2415 - mae: 1.0828 - val_loss: 1.6073 - val_mse: 1.6073 - val_mae: 1.2450\n",
      "Epoch 3/10\n",
      "12/12 [==============================] - 0s 6ms/sample - loss: 1.1859 - mse: 1.1859 - mae: 1.0586 - val_loss: 1.4991 - val_mse: 1.4991 - val_mae: 1.2008\n",
      "Epoch 4/10\n",
      "12/12 [==============================] - 0s 7ms/sample - loss: 1.0946 - mse: 1.0946 - mae: 1.0131 - val_loss: 1.3953 - val_mse: 1.3953 - val_mae: 1.1568\n",
      "Epoch 5/10\n",
      "12/12 [==============================] - 0s 7ms/sample - loss: 0.9628 - mse: 0.9628 - mae: 0.9483 - val_loss: 1.2963 - val_mse: 1.2963 - val_mae: 1.1131\n",
      "Epoch 6/10\n",
      "12/12 [==============================] - 0s 7ms/sample - loss: 0.9030 - mse: 0.9030 - mae: 0.9135 - val_loss: 1.2018 - val_mse: 1.2018 - val_mae: 1.0699\n",
      "Epoch 7/10\n",
      "12/12 [==============================] - 0s 7ms/sample - loss: 0.8351 - mse: 0.8351 - mae: 0.8764 - val_loss: 1.1120 - val_mse: 1.1120 - val_mae: 1.0271\n",
      "Epoch 8/10\n",
      "12/12 [==============================] - 0s 8ms/sample - loss: 0.7665 - mse: 0.7665 - mae: 0.8372 - val_loss: 1.0268 - val_mse: 1.0268 - val_mae: 0.9847\n",
      "Epoch 9/10\n",
      "12/12 [==============================] - 0s 7ms/sample - loss: 0.6971 - mse: 0.6971 - mae: 0.7940 - val_loss: 0.9461 - val_mse: 0.9461 - val_mae: 0.9429\n",
      "Epoch 10/10\n",
      "12/12 [==============================] - 0s 10ms/sample - loss: 0.6355 - mse: 0.6355 - mae: 0.7566 - val_loss: 0.8700 - val_mse: 0.8700 - val_mae: 0.9016\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X_train, y_train, verbose = 1, validation_split = 0.2, batch_size = 52, epochs = 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['loss', 'mse', 'mae', 'val_loss', 'val_mse', 'val_mae'])\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(history.history.keys())\n",
    "#Loss\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc = 'upper left')\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZcAAAEGCAYAAACpXNjrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAAgAElEQVR4nO3de5xV5X3v8c/XAXVMYsYLWhnkQBtCYrSKmSgtnhxrEkGTBuLRRJNUTuIJSY4216IQbdVYI5ZT7TFV87JqxZzESwxBmmCQA9pcjrdRVEQlTDXVGahiBGMiMYC//rGeDZth7z2LYV+H7/v12q+992/dfrNezPxY63nW8ygiMDMzq6Y9Gp2AmZkNPS4uZmZWdS4uZmZWdS4uZmZWdS4uZmZWdcManUCzOPDAA2PMmDGNTsPMrKU8/PDDL0XEiP5xF5dkzJgxdHd3NzoNM7OWIunfS8V9W8zMzKrOxcXMzKrOxcXMzKrOxcXMzKrOxcXMzKrOvcXMzHZTC5b3MXfxKtZs2MjIjnZmTh7PtAmdVdm3i4uZ2W5owfI+Zs9fwcZNWwDo27CR2fNXAFSlwPi2mJnZbmju4lVbC0vBxk1bmLt4VVX27+JiZrYbWrNh407Fd5aLi5nZbmhkR/tOxXeWi4uZ2W5o5uTxtA9v2y7WPryNmZPHV2X/NSsukvaW9KCkxyStlHRxit8k6VlJj6bXUSkuSVdJ6pH0uKSji/Y1XdLq9JpeFH+3pBVpm6skKcX3l7Qkrb9E0n61+jnNzFrRtAmdXHbKEXR2tCOgs6Ody045oiV6i70OnBARv5E0HPiZpLvSspkRcUe/9U8CxqXXscC1wLGS9gcuBLqAAB6WtDAi1qd1ZgD3A4uAKcBdwCxgaUTMkTQrfT+vhj+rmVnLmTahs2rFpL+aXblE5jfp6/D0igqbTAVuTtvdD3RIOgSYDCyJiJdTQVkCTEnL9o2I+yIigJuBaUX7mpc+zyuKm5lZHdS0zUVSm6RHgRfJCsQDadGl6dbXlZL2SrFO4PmizXtTrFK8t0Qc4OCIWAuQ3g+q4o9lZmYDqGlxiYgtEXEUMAo4RtLhwGzgHcB7gP3ZdrtKpXYxiHhukmZI6pbUvW7dup3Z1MzMKqhLb7GI2ADcC0yJiLXp1tfrwD8Dx6TVeoFDizYbBawZID6qRBzghXTbjPT+Ypm8rouIrojoGjFih4nUzMxskGrZW2yEpI70uR14P/B00R99kbWFPJE2WQicmXqNTQReSbe0FgMnStov9fo6EViclr0qaWLa15nAnUX7KvQqm14UNzOzOqhlb7FDgHmS2siK2O0R8UNJyySNILut9SjwubT+IuBkoAd4DfgUQES8LOkS4KG03tcj4uX0+fPATUA7WS+xQm+0OcDtks4CngNOq9lPaWZmO1DW0cq6urqiu7u70WmYmbUUSQ9HRFf/uJ/QNzOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqnNxMTOzqqtZcZG0t6QHJT0maaWki1N8rKQHJK2WdJukPVN8r/S9Jy0fU7Sv2Sm+StLkoviUFOuRNKsoXvIYZmZWHwMWF0mnSHpL+jxL0u2Sjsqx79eBEyLiSOAoYIqkicDlwJURMQ5YD5yV1j8LWB8RbwOuTOsh6TDgdOBdwBTgGkltktqAq4GTgMOAM9K6VDiGmZnVQZ4rl4si4lVJfwr8OXAb8K2BNorMb9LX4ekVwAnAHSk+D5iWPk9N30nL3ydJKX5rRLweEc8CPcAx6dUTEc9ExO+BW4GpaZtyxzAzszrIU1y2pPcPAddExPeBvfLsPF1hPAq8CCwB/g3YEBGb0yq9QGf63Ak8D5CWvwIcUBzvt025+AEVjtE/vxmSuiV1r1u3Ls+PZGZmOeQpLmslXQ18DFiU2i9ytdVExJaIOAoYRXal8c5Sq6V3lVlWrXip/K6LiK6I6BoxYkSpVczMbBCG5Vjno8DJwDcjYr2kkcCsAbbZTkRskHQvMBHokDQsXVmMAtak1XqBQ4FeScOAtwIvF8ULircpFX+pwjHMzMpasLyPuYtXsWbDRkZ2tDNz8nimTSh548MGMOAVSGo3eY7sygOyhvqVA20naYSkjvS5HXg/8BRwD3BqWm06cGf6vDB9Jy1fFhGR4qen3mRjgXHAg8BDwLjUM2xPskb/hWmbcscwMytpwfI+Zs9fQd+GjQTQt2Ejs+evYMHyvkan1pLy9Ba7ALgQuCCF9ga+m2PfhwD3SHqcrBAsiYgfAucBX5HUQ9Y+ckNa/wbggBT/CunqKCJWArcDTwI/Bs5Ot9s2A+cAi8mK1u1pXSocw8yspLmLV7Fx05btYhs3bWHu4lUNyqi15bktdiowAXgEICL6JO070EYR8Xjarn/8GbZdBRXHfwecVmZflwKXlogvAhblPYaZWTlrNmzcqbhVlqdh/vV0qykAJO1T25TMzOpvZEf7TsWtsjzFZX7qLfZWSZ8C7gZurG1aZmb1NXPyeNqHt20Xax/exszJ4xuUUWsb8LZYRFwu6STg98CRwKURcVfNMzMzq6NCrzD3FquOPG0uACuA30XEPWnMsDdFxG9rmZiZWb1Nm9DpYlIleXqLfZqsO/D1KTQad+01M7MK8rS5fIHs4cdfA0TEL4CDapmUmZm1tjzF5XdpYEggGy+M0kOsmJmZAfmKy88lnQvsLenPyEZF/mFt0zIzs1aWp7icC7wKPA18EVgKnF/LpMzMrLXl6S32AeBbEXFtrZMxM7OhIc+Vy/8AVkv6hqRxNc7HzMyGgDyjIp8OdAF9wC2Sfirp05LeVPPszMysJeWd9GsD2UjIN5E953IG8Jik/1W71MzMrFXleYjyJEnfA34KvAWYGBEfIBsK5rwa52dmZi0oT4P+XwDXRsSy4mBE/FbSZ2qTlpmZtbI8A1d+XNKBkqakUHdEvJSW3V3T7MzMrCUNWFwknQL8A9ltMQHfkvTliPhBrZMzs92L57AfOvLcFrsQeE9EvAAg6WCyOV1cXMysagpz2BemGi7MYQ+4wLSgPL3F9igUlmRdzu3MzHLzHPZDS54icbekRZI+KemTZMPvLx5oI0mHSrpH0lOSVkr6YopfJKlP0qPpdXLRNrMl9UhaJWlyUXxKivVImlUUHyvpAUmrJd0mac8U3yt970nLx+Q+I2bWEJ7DfmjJU1z+CpgHHAMcmz7PzLHdZuCrEfFOsiH7z5Z0WFp2ZUQclV6LANKy04F3AVOAayS1pVGYrwZOAg4Dzijaz+VpX+OA9cBZKX4WsD4i3gZcmdYzsybmOeyHljxP6EdE3BYRX4iIv4yI70VE5NhubUQ8kj6/CjwFVLpxOhW4NSJej4hngR6ygnYM0BMRz6Sh/28FpkoScAJwR9p+HjCtaF/z0uc7gPel9c2sSXkO+6GlbHGRtF7SyyVe6yW9vDMHSbelJgAPpNA5kh6XdKOk/VKsE3i+aLPeFCsXPwDYEBGb+8W321da/kpav39eMyR1S+pet27dzvxIZlZl0yZ0ctkpR9DZ0Y6Azo52LjvlCDfmt6hKvcUOrMYBJL0Z+D7wpYj4taRrgUuASO9/D3ya0hOQBaULYFRYnwGWbQtEXAdcB9DV1TXg1ZiZ1ZbnsB86yhaXiNjabUPSHwPHkf2B/llErMizc0nDyQrLdyJiftrvC0XL/4ltE4/1AocWbT4KWJM+l4q/BHRIGpauTorXL+yrV9Iw4K3ATl1tmZnZ4OUZW+x84BayW02jyEZGnp1jOwE3AE9FxBVF8UOKVvsI8ET6vBA4PfX0GguMAx4EHgLGpZ5he5I1+i9M7T73AKem7acDdxbta3r6fCqwLE87kZmZVUeehyg/Cbw7Il4DkHQp8DBw2QDbTSIbl2yFpEdT7Gtkvb2OIrsK+iXwWYCIWCnpduBJsp5mZxeuniSdQ9b9uQ24MSJWpv2dB9wq6W+B5WTFjPT+bUk9ZFcsp+f4Oc3MrEo00H/oJf0Y+GhE/Dp93xe4JSI+WIf86qarqyu6u7sbnYaZWUuR9HBEdPWP57lyeQ1YKWkx2dXGicDPJF0BEBFfqWqmZmbW8vIUlx+lV8H9NcrFzMyGiDxD7t8w0DpmZmbF8vQWmyLpIUkvDvYhSjMz273kuS32j8BHgRXAG7VNx8zMhoI8xaUXeDQiXFjMzCyXPMXlXOBfJN0LvF4IRsRVtUrKzMxaW57icjGwCejAt8XMzCyHPMXloIh4d80zMTOzISPPZGFLJZ1Q80zMzGzIyFNcPgP8P0m/cVdkMzPLI89tsarM62JmZruPSjNRngFb53V5T0RsKbyAGfVK0MzMWk+l22Iziz5f02/ZZ2qQi5mZDRGViovKfC713czMbKtKxSXKfC713czMbKtKDfrvkPQI2VXK+PSZ9P3tNc/MzMxaVqXickTdsjAzsyGlbHGJiH+rZyJmZjZ05HmIclAkHSrpHklPSVop6Yspvr+kJZJWp/f9UlySrpLUI+lxSUcX7Wt6Wn+1pOlF8XdLWpG2uUqSKh3DzMzqo2bFBdgMfDUi3glMBM6WdBgwC1gaEeOApek7wEnAuPSaAVwLWaEALgSOBY4BLiwqFtemdQvbTUnxcscwM7M6qPQQ5d3p/RuD2XFErI2IR9LnV4GngE5gKjAvrTYPmJY+TwVujsz9QIekQ4DJwJKIeDki1gNLgClp2b4RcV9EBHBzv32VOoaZmdVBpQb9UZImAR+RdAv9nm2JiMfzHkTSGGAC8ABwcESsTftYK+mgtFon8HzRZr0pVineWyJOhWP0z2sGabSB0aNH5/1xzMxsAJWKy8XARcAodnxCP4D35jmApDcD3we+FBG/Ts0iJVctEYtBxHOLiOuA6wC6urr87I6ZWZVU6i12G3CbpIsj4sLB7FzScLLC8p2ImJ/CL0g6JF1RHAK8mOK9wKFFm48C1qT48f3i96b4qBLrVzqGmZnVwYAN+hFxoaSTJc1JrykDbQNZ7y/gBuCpiLiiaNFCoNDjazpwZ1H8zNRrbCLwSrq1tRg4UdJ+qSH/RGBxWvaqpInpWGf221epY5i1vAXL+5g0ZxljZ/2ISXOWsWB5X6NTMtvBgEPuS/pb4Djguyl0rqTjIuKCATadBPwFsELSoyn2NWAOcLuks4DngNPSskXAyUAP8BrwKYCIeFnSJcBDab2vR0RhPpnPAzcB7cBd6UWFY5i1tAXL+5g9fwUbN20BoG/DRmbPXwHAtAmdlTY1qytlHa0qrCA9DkxIQ+0jaRjwSET8cR3yq5uurq7o7u5udBpmFU2as4y+DRt3iHd2tPPzWZ4w1upP0sMR0dU/nvc5l32LPr+lOimZ2c5aU6KwVIqbNUqemSj/DnhE0lKyHlrHA39Ty6TMrLSRHe0lr1xGdrQ3IBuz8vI06P9fsjaXRen13oj4Tq0TM7MdzZw8nvbhbdvF2oe3MXPy+AZlZFZanisXIqIPmD/gimZWU4VG+7mLV7Fmw0ZGdrQzc/J4N+Zb08lVXMyseUyb0OliYk2vlgNXmpnZbqpicZHUJumxeiVjZmZDQ8Xikp5teVKSr8HNzCy3PG0uBwJPSboP+G0hGBGn1CwrMzNraXmKy5yaZ2FmZkPKgMUlIpZKGgWMi4h7JO0NtA20nZmZ7b4G7C0m6dNkowxfn0Kj8SjDZmZWQZ6uyF8AJgK/BoiIXwAH1zIpMzNrbXmKy+8i4veFL5J8S8zMzCrKU1x+LulcYG9JfwbcBvywtmmZmVkry1NczgVeBZ4GvggsBc6vZVJmZtba8vQW2yLpeuBfgQBWR8QbNc/MzMxaVp5pjqcA15FNFyxglKTPRMTdtU7OzMxaU56HKP8BeH/qJYakt5N1RX5nLRMzM7PWlafN5cVCYYGtXZHXDbSRpBslvSjpiaLYRZL6JD2aXicXLZstqUfSKkmTi+JTUqxH0qyi+FhJD0haLek2SXum+F7pe09aPibHz2g2oAXL+5g0ZxljZ/2ISXOWsWB5X6NTMmtaZYuLpA9L+jDwhKSFkj4p6ROSFgAP5tj3TcCUEvErI+Ko9FqUjnUYcDrwrrTNNWlE5jbgauAk4DDgjLQuwOVpX+OA9cBZKX4WsD4i3gZcmdYz2yULlvcxe/4K+jZsJIC+DRuZPX+FC4xZGZVui51W9PkVoHA18Spw0EA7joif7MRVw1Tg1oh4HXhWUg9wTFrWExHPAEi6FZgq6SngBODjaZ15wEXAtWlfF6X4HcA/SlJERM5czICsoBRmfNxDYku/f0IbN21h7uJVnrjLrISyxSUi/qJGxzxH0plAN/DViFgPdAL3F63Tm2IAz/eLHwscAGyIiM0l1u8sbBMRmyW9ktZ/qX8ikmYAMwBGjx696z+ZDRmFK5WNm7YA7FBYCtZs2FjPtMxaRp7eYqOBc4AxxesPcsj9a4FLyLo0XwL8PfBpsl5o/QWlb9tFhfUZYNn2wYjryHrC0dXV5Ssb23q10pezaIzsaK9xRmatKU9vsYXAzcASYJeeb4mIFwqfJf0T25707wUOLVp1FLAmfS4VfwnokDQsXb0Ur1/YV6+kYcBbgZd3JW8b+hYs7+Pif1nJ+tc25d6mfXgbMyePr2FWZq0rT3H5fURcUY2DSTokItamrx8BCj3JFgLflXQFMBIYR9ZpQMA4SWOBPrJG/49HREi6BzgVuBWYzraRmhem7/el5cvc3mKV9L8FVkmbxBsRjOxoZ+bk8W5vMSsjT3H5pqQLgMXA64VgRDxeaSNJtwDHAwdK6gUuBI6XdBTZbapfAp9N+1op6XbgSWAzcHaaYhlJ56RjtwE3RsTKdIjzgFsl/S2wHLghxW8Avp06BbxMVpDMypq7eFWuwtI+vI3LTjnCBcUsBw30n3pJlwD/E3iGbbfFIiLeW+Pc6qqrqyu6u7sbnYY1wNhZPyrdKFek01cqZiVJejgiuvrH81y5fBQYk7oJmw05Izvayzbg+2rFbHDyPKH/OPCWWidi1igzJ4+nffiO0xR1tA93YTEbpDxXLgcAT0t6gO3bXAbTFdms6RSKR+GBSTfWm+26PMXl0ppnYdZg0yZ0upiYVVGe+VyW1iMRMzMbOvI8of8q255wH0bWJfj1iNi3lomZmVnrynPlsrUxX9IewCnAkbVMyszMWlue3mJbRcQbEXEH8IEa5WNmZkNAnttiHy76ugfQRenBIc3MzIB8vcWK53XZTDZsy9SaZGNmZkNCnjaXWs3rYjYoC5b3cdHClWzYmI1gvN8+w7nwz9/lrsRmTaRscZH0tQrbRURcVoN8zCpasLyPmd97jE1vbBsNbP1rm5h5x2MALjBmTaJSg/6WEq89yUYy/uvap2a2o7mLV21XWAo2bQnmLl7VgIzMrJRK0xxfXvgs6c1ks1GeSTYv/dzap2a2o0rTCnvKYbPmUbErsqQOSRcBK4A3A++JiK9GxH/UIzmz/ipNK+wph82aR9niIuky4GFgE3BkRFwQEb+qW2ZmJcycPJ7he+zYE354mzzlsFkTqXTlch7wB8BfAb+U9HJ6rZfkOemtIaZN6GTuaUfS0T58a2y/fYYz99Qj3Zhv1kQqdUUeXmGZWcN4BGOz5lepQX/gScXNdpGfWTEbmnZqbLGdIelGSS9KeqIotr+kJZJWp/f9UlySrpLUI+lxSUcXbTM9rb9a0vSi+LslrUjbXCVJlY5hzafwzEqhsMC2Z1YWLO9rYGZmtqtqVlyAm4Ap/WKzgKURMQ5Ymr4DnASMS68ZwLWQFQrgQuBY4BjgwqJicW1at7DdlAGOYU3Gz6yYDV01Ky4R8ROgf8P/VGBe+jwPmFYUvzky9wMdkg4BJgNLIuLliFgPLAGmpGX7RsR9ERHAzf32VeoY1mT8zIrZ0FVp+Jf1bJskbLtFZMO/7D+I4x0cEWvJdrBW0kEp3gk8X7Reb4pViveWiFc6xo4/iDSD7OqH0aNHD+LHsV0xsqOdvjJFxM+smLW2SlcuBwIjSrwK8WoqNYR/DCK+UyLiuojoioiuESOq/SNZwYLlfUyas4yxs37EpDnLtran+JkVs6Erd2+x1P6xd1FozSCO94KkQ9IVxSHAiyneCxxatN6otP9e4Ph+8XtTfFSJ9SsdwxpgwfI+Zs9fwcZN2T+nvg0bmT1/BbBtkEn3FjMbevJMFvZB4EqyP+C/Irv99AvgHYM43kJgOjAnvd9ZFD9H0q1kjfevpOKwGPhGUSP+icDsiHhZ0quSJgIPkI159s0BjmENMHfxqq2FpWDjpi3MXbxq6/MqLiRmQ0+eycIuBSYBd0fEBEkfAP77QBtJuoXsquNASb1kvb7mALdLOgt4jm0TkS0CTgZ6gNeATwGkInIJ8FBa7+sRUegk8HmyHmntwF3pRYVjWAOUa5h3g73Z0JanuGyOiHWS9pCkiFgi6dKBNoqIM8osel+JdQM4u8x+bgRuLBHvBg4vEf9VqWNYY5RrtHeDvdnQlqe4vCLpTcDPgJslvQi8Udu0rJUsWN7H3MWrWLNhIyM72pk5efzWW10zJ4/frs0FoH14mxvszYa4PMVlGvA74EtkbRtvBT5Uy6SsdQzUYF8oMuWKj5kNTcruSFVYQfpGRHxtoFir6+rqiu7u7kan0XImzVlW8rZXZ0c7P591QgMyMrN6kvRwRHT1j+d5Qr//EC4AH9z1lGwocIO9mZVS6Qn9zwKfA94u6ZGiRW8B/F98A9xgb2alVbpyuZ2sG++i9F54TarQE8x2MzMnj6d9eNt2MTfYm1mlJ/TXA+uB0yQdDhyXFv0UP/VuiRvszayUPE/on032DMqCFLpd0tURcU1NM7OW4afszay/PF2RPwscExG/gaynGPD/ARcXMzMrKU9vMQGbir5vovSoxGZmZkDl3mLDImIz8G3gfknfT4s+wrbJuGyIqPSUvZnZzqp0W+xB4OiI+DtJ9wD/leyK5XMR8VCF7azF5BkW38xsZ1QqLltvfaVi4oIyhBRfqewhsaXfSA3Fw+Kbme2sSsVlhKSvlFsYEVfUIB+rg/5XKv0LS4GfsjezwapUXNqAN+PG+yFlwfI+vnr7Y2ULSjE/ZW9mg1WpuKyNiK/XLROrucIVS57C4qfszWxX5GpzsaGh1JTDxdok3ohwbzEz22WViotncxxiKrWhtA9v47JTjnBBMbOqKPsQZdFc9TZElGtDaZNcWMysqvI8oV91kn4paYWkRyV1p9j+kpZIWp3e90txSbpKUo+kxyUdXbSf6Wn91ZKmF8Xfnfbfk7b1LT7Kj2D89x890oXFzKqqIcUl+bOIOKpoBrNZwNKIGAcsTd8BTgLGpdcM4FrIihFwIXAscAxwYaEgpXVmFG1XasKz3c60CZ1cdsoRdHa0I7LZIn3FYma1kGfgynqZChyfPs8D7gXOS/GbI5uP+X5JHZIOSesuKdy+k7QEmCLpXmDfiLgvxW8GpgF31e0naWIewdjM6qFRVy4B3C3pYUkzUuzgiFgLkN4PSvFO4PmibXtTrFK8t0R8B5JmSOqW1L1u3bpd/JHMzKygUVcukyJijaSDgCWSnq6wbqn2khhEfMdgxHXAdQBdXV0DP/xhZma5NKS4RMSa9P6ipB+QtZm8IOmQiFibbnsVZrvsBQ4t2nwUsCbFj+8XvzfFR5VYf0jw6MVm1grqfltM0pskvaXwGTgReAJYCBR6fE0H7kyfFwJnpl5jE4FX0m2zxcCJkvZLDfknAovTslclTUy9xM4s2ldLu2DBCr5826P0bdhIsG304gXL+xqdmpnZdhpx5XIw8IPUO3gY8N2I+LGkh8imUD4LeA44La2/CDgZ6AFeAz4F2XM4ki5h22jNXy96NufzwE1AO1lDfss35i9Y3sd37n9uh/t7Hr3YzJpR3YtLRDwDHFki/itKjAqQeomdXWZfNwI3loh3A4fvcrJNZO7iVaUbjvDoxWbWfBr5nIvthEoFxKMXm1mzcXFpEeUKiMCjF5tZ03FxaSILlvcxac4yxs76EZPmLNuuob7U0C0CPjFxtNtbzKzpNNMT+ru1geaxLxQQd0M2s1bg4tIkSs210r8nmIduMbNW4dtiTaJcg717gplZK/KVS52Ve8J+ZEc7fSUKiXuCmVkr8pVLHRXaVUo9YV9urhX3BDOzVuTiUkcDtat4rhUzGyp8W6yOBmpXcYO9mQ0VvnKpo3LtJ25XMbOhxsWljtyuYma7C98WqyM/CGlmuwsXlzpzu4qZ7Q58W8zMzKrOVy67wFMOm5mV5uIySAMNNGlmtjvzbbFBqvRApJnZ7s7FZZA80KSZWXlDtrhImiJplaQeSbOqvX8/EGlmVt6QLC6S2oCrgZOAw4AzJB1WzWP4gUgzs/KGaoP+MUBPRDwDIOlWYCrwZLUO4AcizczKG6rFpRN4vuh7L3Bs/5UkzQBmAIwePXqnD+IHIs3MShuSt8UAlYjFDoGI6yKiKyK6RowYUYe0zMx2D0O1uPQChxZ9HwWsaVAuZma7naFaXB4CxkkaK2lP4HRgYYNzMjPbbQzJNpeI2CzpHGAx0AbcGBErG5yWmdluY0gWF4CIWAQsanQeZma7I0Xs0M69W5K0Dvj3AVY7EHipDulUk3OuD+dcH865PnYm5/8SETv0iHJx2QmSuiOiq9F57AznXB/OuT6cc31UI+eh2qBvZmYN5OJiZmZV5+Kyc65rdAKD4JzrwznXh3Ouj13O2W0uZmZWdb5yMTOzqnNxMTOzqnNxyanWk49Vg6RDJd0j6SlJKyV9McX3l7RE0ur0vl+jcy0mqU3Sckk/TN/HSnog5XtbGsKnqUjqkHSHpKfT+f6TFjjPX07/Lp6QdIukvZvtXEu6UdKLkp4oipU8r8pclX4nH5d0dBPlPDf923hc0g8kdRQtm51yXiVpcrPkXLTsrySFpAPT90GdZxeXHOox+ViVbAa+GhHvBCYCZ6c8ZwFLI2IcsDR9byZfBJ4q+n45cGXKdz1wVkOyquz/AD+OiHcAR5Ll37TnWVIn8AWgKyIOJxsW6XSa71zfBEzpFyt3Xk8CxqXXDODaOuXY303smPMS4PCI+GPgF8BsgPT7eDrwrrTNNRGGtAMAAAYESURBVOnvS73dxI45I+lQ4APAc0XhQZ1nF5d8tk4+FhG/BwqTjzWViFgbEY+kz6+S/cHrJMt1XlptHjCtMRnuSNIo4IPA9em7gBOAO9IqTZUvgKR9gfcCNwBExO8jYgNNfJ6TYUC7pGHAPsBamuxcR8RPgJf7hcud16nAzZG5H+iQdEh9Mt2mVM4RcXdEbE5f7ycbmR2ynG+NiNcj4lmgh+zvS12VOc8AVwLnsv0UJYM6zy4u+ZSafKypZwmTNAaYADwAHBwRayErQMBBjctsB/9A9o/5jfT9AGBD0S9mM57rPwTWAf+cbuddL+lNNPF5jog+4H+T/Y90LfAK8DDNf66h/Hltld/LTwN3pc9Nm7OkDwN9EfFYv0WDytnFJZ9ck481C0lvBr4PfCkift3ofMqR9CHgxYh4uDhcYtVmO9fDgKOBayNiAvBbmugWWCmpnWIqMBYYCbyJ7HZHf812ritp+n8rks4nu139nUKoxGoNz1nSPsD5wN+UWlwiNmDOLi75tMzkY5KGkxWW70TE/BR+oXAZm95fbFR+/UwCPizpl2S3Gk8gu5LpSLduoDnPdS/QGxEPpO93kBWbZj3PAO8Hno2IdRGxCZgP/CnNf66h/Hlt6t9LSdOBDwGfiG0PFDZrzn9E9h+Px9Lv4yjgEUl/wCBzdnHJpyUmH0vtFTcAT0XEFUWLFgLT0+fpwJ31zq2UiJgdEaMiYgzZOV0WEZ8A7gFOTas1Tb4FEfEfwPOSxqfQ+4AnadLznDwHTJS0T/p3Usi5qc91Uu68LgTOTL2ZJgKvFG6fNZqkKcB5wIcj4rWiRQuB0yXtJWksWSP5g43IsVhErIiIgyJiTPp97AWOTv/WB3eeI8KvHC/gZLJeH/8GnN/ofMrkeBzZ5erjwKPpdTJZO8ZSYHV637/RuZbI/Xjgh+nzH5L9wvUA3wP2anR+JfI9CuhO53oBsF+zn2fgYuBp4Ang28BezXaugVvI2oQ2pT9wZ5U7r2S3a65Ov5MryHrCNUvOPWTtFIXfw28VrX9+ynkVcFKz5Nxv+S+BA3flPHv4FzMzqzrfFjMzs6pzcTEzs6pzcTEzs6pzcTEzs6pzcTEzs6pzcTEbgKTfVFh2vNJozo0m6VJJz/fPNz1TcVsa1faBNDRQYVnJEXrVAqOAW3NzcTFrQWVG0v0XSg+CeBawPiLeRjYw4eVpHyVH6G2hUcCtibm4mOWQnk6eq2wulBWSPla0eN80Z8eTkr4laY/0R/qmovW/XGKfN6X1fyrpF2mstcL8NnMlPZTmz/hsih+vbL6e75I9zLadiLg/Sj85XTyq8B3A+9JT+uVG6G2JUcCtuQ0beBUzA04heyr/SOBA4CFJP0nLjiH7H/6/Az9O6z4LdEY2dwoqmiyqnzHAfyMb2+keSW8DziQbYuM9kvYCfi7p7qJjHZ6KQV5bR7WNiM2SXiF76r2TbDj4guLRbvuPgnvsThzPzFcuZjkdB9wSEVsi4gXgX4H3pGUPpv/lbyEbVuM44BngDyV9M40zVW506tsj4o2IWJ22eQdwItlYTo+STZlwANkYVIVj7UxhgfKj2u5s3Cw3FxezfEr9wS3o/4c3ImI92VXOvcDZpMnQ8mybjvWXEXFUeo2NiMKVy293Lm2gaFTbNALyW8kmiio32m2zjtxrLcTFxSyfnwAfS+0hI8hmoiyMZntMGjF7D+BjwM+UzT++R0R8H/hrsiH5SzkttdH8EdkgkquAxcDn0/QJSHp7moxssIpHFT6VbPTpoPwIvS0xCrg1N7e5mFWQ/qf/OvAD4E+Ax8iuLs6NiP+Q9A7gPmAOcARZEfpB+vzPqeBAmkO9hFVkt9gOBj4XEb+TdD1ZW8wjqeF9HTmmH5b0d8DHgX0k9QLXR8RFZNMwfFtSD9kVy+kAEbFS0u1kQ+9vBs5Ot/aQdA5ZkWsDboyIlTlOl9lWHhXZrAJJRwL/FBFVn+dc0k1k0wzcMdC6Zq3Gt8XMypD0ObIG+gsanYtZq/GVi5mZVZ2vXMzMrOpcXMzMrOpcXMzMrOpcXMzMrOpcXMzMrOr+Ez6kIm72c2poAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x, y = loadData1()\n",
    "plotData(x[:, 0], y, 'Jobs per 1000', 'Total Number of Employees')\n",
    "plt.figure()\n",
    "plotData(x[:, 1], y, 'Location Variable', 'Total Number of Employees')\n",
    "plt.figure()\n",
    "plotData(x[:, 2], y, 'Hourly Wage Average', 'Total Number of Employees')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 [-153659.12]\n",
      "2 [-152885.31]\n",
      "3 [-152953.92]\n",
      "4 [-153246.5]\n",
      "5 [-153228.31]\n",
      "6 [-153693.02]\n",
      "7 [-153623.88]\n",
      "8 [-153532.16]\n",
      "9 [-153519.25]\n",
      "10 [-153489.5]\n",
      "11 [-153760.23]\n",
      "12 [-153697.88]\n",
      "13 [-153067.06]\n",
      "14 [-153532.34]\n",
      "15 [-153619.11]\n",
      "16 [-152902.28]\n",
      "17 [-152947.05]\n",
      "18 [-153996.02]\n",
      "19 [-153084.6]\n",
      "20 [-153528.69]\n",
      "21 [-153958.64]\n",
      "22 [-153614.16]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x, y = loadData1()\n",
    "x = np.expand_dims(x, axis = 1)\n",
    "predictions = []\n",
    "for i in range(len(x)):\n",
    "    x[i]= scaler_x.transform(x[i].reshape(1,-1))\n",
    "    ynew= model.predict(x[i])\n",
    "    #invert normalize\n",
    "    ynew = scaler_y.inverse_transform(ynew.reshape(1,-1))\n",
    "    x[i] = scaler_x.inverse_transform(x[i])\n",
    "    predictions.append(ynew[0])\n",
    "##print(\"X=%s, Predicted=%s\" % (Xnew[0], ynew[0]))\n",
    "plt.figure()\n",
    "plotData(y, predictions, 'measured', 'predicted' )\n",
    "for i in range(len(predictions)):\n",
    "    print(i+1, predictions[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdin",
     "output_type": "stream",
     "text": [
      "Enter the jobs per 1000 people normalized:  44.3\n",
      "Enter the location quotient normalized:  .85\n",
      "Enter the average hourly wage normalzied:  63.35\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X:  [44.3   0.85 63.35] Predicted =  [-153660.94]\n"
     ]
    }
   ],
   "source": [
    "pop = float(input(\"Enter the jobs per 1000 people normalized: \"))\n",
    "dense = float(input(\"Enter the location quotient normalized: \"))\n",
    "days = float(input(\"Enter the average hourly wage normalzied: \"))\n",
    "x_spec = [[pop, dense, days]]\n",
    "x_spec = scaler_x.transform(x_spec)\n",
    "y_hat = model.predict(x_spec)\n",
    "y_hat = scaler_y.inverse_transform(y_hat.reshape(-1,1))\n",
    "x_spec = scaler_x.inverse_transform(x_spec)\n",
    "print('X: ',x_spec[0], 'Predicted = ', y_hat[0])    #Is there a way to un normalize data quickly?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}