chapter03_introduction-to-keras-and-tf.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "This is a companion notebook for the book [Deep Learning with Python, Second Edition](https://www.manning.com/books/deep-learning-with-python-second-edition?a_aid=keras&a_bid=76564dff). For readability, it only contains runnable code blocks and section titles, and omits everything else in the book: text paragraphs, figures, and pseudocode.\n",
    "\n",
    "**If you want to be able to follow what's going on, I recommend reading the notebook side by side with your copy of the book.**\n",
    "\n",
    "This notebook was generated for TensorFlow 2.6."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "# Introduction to Keras and TensorFlow"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "## What's TensorFlow?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "## What's Keras?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "## Keras and TensorFlow: A brief history"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "## Setting up a deep-learning workspace"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "### Jupyter notebooks: The preferred way to run deep-learning experiments"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "### Using Colaboratory"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "#### First steps with Colaboratory"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "#### Installing packages with pip"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "#### Using the GPU runtime"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "## First steps with TensorFlow"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "#### Constant tensors and variables"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**All-ones or all-zeros tensors**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tf.Tensor(\n",
      "[[1.]\n",
      " [1.]], shape=(2, 1), dtype=float32)\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "x = tf.ones(shape=(2, 1))\n",
    "print(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tf.Tensor(\n",
      "[[0.]\n",
      " [0.]], shape=(2, 1), dtype=float32)\n"
     ]
    }
   ],
   "source": [
    "x = tf.zeros(shape=(2, 1))\n",
    "print(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Random tensors**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tf.Tensor(\n",
      "[[-1.0450664 ]\n",
      " [-0.3079132 ]\n",
      " [ 0.03667038]], shape=(3, 1), dtype=float32)\n"
     ]
    }
   ],
   "source": [
    "x = tf.random.normal(shape=(3, 1), mean=0., stddev=1.)\n",
    "print(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tf.Tensor(\n",
      "[[0.9446955 ]\n",
      " [0.3969561 ]\n",
      " [0.50828195]], shape=(3, 1), dtype=float32)\n"
     ]
    }
   ],
   "source": [
    "x = tf.random.uniform(shape=(3, 1), minval=0., maxval=1.)\n",
    "print(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab_type": "text"
   },
   "outputs": [
    {
     "ename": "SyntaxError",
     "evalue": "invalid syntax (<ipython-input-6-9cca9502700c>, line 1)",
     "output_type": "error",
     "traceback": [
      "\u001b[1;36m  File \u001b[1;32m\"<ipython-input-6-9cca9502700c>\"\u001b[1;36m, line \u001b[1;32m1\u001b[0m\n\u001b[1;33m    **NumPy arrays are assignable**\u001b[0m\n\u001b[1;37m    ^\u001b[0m\n\u001b[1;31mSyntaxError\u001b[0m\u001b[1;31m:\u001b[0m invalid syntax\n"
     ]
    }
   ],
   "source": [
    "**NumPy arrays are assignable**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "x = np.ones(shape=(2, 2))\n",
    "x[0, 0] = 0."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Creating a TensorFlow variable**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<tf.Variable 'Variable:0' shape=(3, 1) dtype=float32, numpy=\n",
      "array([[ 1.6678083],\n",
      "       [-1.0415488],\n",
      "       [-1.6678433]], dtype=float32)>\n"
     ]
    }
   ],
   "source": [
    "v = tf.Variable(initial_value=tf.random.normal(shape=(3, 1)))\n",
    "print(v)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab_type": "text"
   },
   "outputs": [
    {
     "ename": "SyntaxError",
     "evalue": "invalid syntax (<ipython-input-9-e09799acdc58>, line 1)",
     "output_type": "error",
     "traceback": [
      "\u001b[1;36m  File \u001b[1;32m\"<ipython-input-9-e09799acdc58>\"\u001b[1;36m, line \u001b[1;32m1\u001b[0m\n\u001b[1;33m    **Assigning a value to a TensorFlow variable**\u001b[0m\n\u001b[1;37m    ^\u001b[0m\n\u001b[1;31mSyntaxError\u001b[0m\u001b[1;31m:\u001b[0m invalid syntax\n"
     ]
    }
   ],
   "source": [
    "**Assigning a value to a TensorFlow variable**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Variable 'UnreadVariable' shape=(3, 1) dtype=float32, numpy=\n",
       "array([[1.],\n",
       "       [1.],\n",
       "       [1.]], dtype=float32)>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "v.assign(tf.ones((3, 1)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Assigning a value to a subset of a TensorFlow variable**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Variable 'UnreadVariable' shape=(3, 1) dtype=float32, numpy=\n",
       "array([[3.],\n",
       "       [1.],\n",
       "       [1.]], dtype=float32)>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "v[0, 0].assign(3.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Using `assign_add`**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Variable 'UnreadVariable' shape=(3, 1) dtype=float32, numpy=\n",
       "array([[4.],\n",
       "       [2.],\n",
       "       [2.]], dtype=float32)>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "v.assign_add(tf.ones((3, 1)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab_type": "text"
   },
   "outputs": [],
   "source": [
    "#### Tensor operations: Doing math in TensorFlow"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**A few basic math operations**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "a = tf.ones((2, 2))\n",
    "b = tf.square(a)\n",
    "c = tf.sqrt(a)\n",
    "d = b + c\n",
    "e = tf.matmul(a, b)\n",
    "e *= d"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "#### A second look at the GradientTape API"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Using the `GradientTape`**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "input_var = tf.Variable(initial_value=3.)\n",
    "with tf.GradientTape() as tape:\n",
    "   result = tf.square(input_var)\n",
    "gradient = tape.gradient(result, input_var)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Using `GradientTape` with constant tensor inputs**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "input_const = tf.constant(3.)\n",
    "with tf.GradientTape() as tape:\n",
    "   tape.watch(input_const)\n",
    "   result = tf.square(input_const)\n",
    "gradient = tape.gradient(result, input_const)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Using nested gradient tapes to compute second-order gradients**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "time = tf.Variable(0.)\n",
    "with tf.GradientTape() as outer_tape:\n",
    "    with tf.GradientTape() as inner_tape:\n",
    "        position =  4.9 * time ** 2\n",
    "    speed = inner_tape.gradient(position, time)\n",
    "acceleration = outer_tape.gradient(speed, time)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "#### An end-to-end example: A linear classifier in pure TensorFlow"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Generating two classes of random points in a 2D plane**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "num_samples_per_class = 1000\n",
    "negative_samples = np.random.multivariate_normal(\n",
    "    mean=[0, 3],\n",
    "    cov=[[1, 0.5],[0.5, 1]],\n",
    "    size=num_samples_per_class)\n",
    "positive_samples = np.random.multivariate_normal(\n",
    "    mean=[3, 0],\n",
    "    cov=[[1, 0.5],[0.5, 1]],\n",
    "    size=num_samples_per_class)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Stacking the two classes into an array with shape (2000, 2)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "inputs = np.vstack((negative_samples, positive_samples)).astype(np.float32)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Generating the corresponding targets (0 and 1)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "targets = np.vstack((np.zeros((num_samples_per_class, 1), dtype=\"float32\"),\n",
    "                     np.ones((num_samples_per_class, 1), dtype=\"float32\")))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Plotting the two point classes**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab_type": "code"
   },
   "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"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.scatter(inputs[:, 0], inputs[:, 1], c=targets[:, 0])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Creating the linear classifier variables**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "input_dim = 2\n",
    "output_dim = 1\n",
    "W = tf.Variable(initial_value=tf.random.uniform(shape=(input_dim, output_dim)))\n",
    "b = tf.Variable(initial_value=tf.zeros(shape=(output_dim,)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**The forward pass function**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "def model(inputs):\n",
    "    return tf.matmul(inputs, W) + b"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**The mean squared error loss function**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "def square_loss(targets, predictions):\n",
    "    per_sample_losses = tf.square(targets - predictions)\n",
    "    return tf.reduce_mean(per_sample_losses)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**The training step function**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "learning_rate = 0.1\n",
    "\n",
    "def training_step(inputs, targets):\n",
    "    with tf.GradientTape() as tape:\n",
    "        predictions = model(inputs)\n",
    "        loss = square_loss(predictions, targets)\n",
    "    grad_loss_wrt_W, grad_loss_wrt_b = tape.gradient(loss, [W, b])\n",
    "    W.assign_sub(grad_loss_wrt_W * learning_rate)\n",
    "    b.assign_sub(grad_loss_wrt_b * learning_rate)\n",
    "    return loss"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**The batch training loop**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss at step 0: 2.0009\n",
      "Loss at step 1: 0.1622\n",
      "Loss at step 2: 0.1006\n",
      "Loss at step 3: 0.0874\n",
      "Loss at step 4: 0.0811\n",
      "Loss at step 5: 0.0761\n",
      "Loss at step 6: 0.0717\n",
      "Loss at step 7: 0.0677\n",
      "Loss at step 8: 0.0640\n",
      "Loss at step 9: 0.0606\n",
      "Loss at step 10: 0.0575\n",
      "Loss at step 11: 0.0547\n",
      "Loss at step 12: 0.0521\n",
      "Loss at step 13: 0.0498\n",
      "Loss at step 14: 0.0476\n",
      "Loss at step 15: 0.0457\n",
      "Loss at step 16: 0.0438\n",
      "Loss at step 17: 0.0422\n",
      "Loss at step 18: 0.0407\n",
      "Loss at step 19: 0.0393\n",
      "Loss at step 20: 0.0380\n",
      "Loss at step 21: 0.0369\n",
      "Loss at step 22: 0.0358\n",
      "Loss at step 23: 0.0349\n",
      "Loss at step 24: 0.0340\n",
      "Loss at step 25: 0.0332\n",
      "Loss at step 26: 0.0324\n",
      "Loss at step 27: 0.0318\n",
      "Loss at step 28: 0.0311\n",
      "Loss at step 29: 0.0306\n",
      "Loss at step 30: 0.0301\n",
      "Loss at step 31: 0.0296\n",
      "Loss at step 32: 0.0292\n",
      "Loss at step 33: 0.0288\n",
      "Loss at step 34: 0.0284\n",
      "Loss at step 35: 0.0281\n",
      "Loss at step 36: 0.0278\n",
      "Loss at step 37: 0.0275\n",
      "Loss at step 38: 0.0272\n",
      "Loss at step 39: 0.0270\n"
     ]
    }
   ],
   "source": [
    "for step in range(40):\n",
    "    loss = training_step(inputs, targets)\n",
    "    print(f\"Loss at step {step}: {loss:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAD4CAYAAADxeG0DAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAB6NklEQVR4nO2dd3gUVReH3zvbNz0h9CpNAQUVRRQrFhAbdrErIopd7PjZu6KIHbEDCnYUFRW7iFKsqFTpPT3ZPvf7YzYhm53dbJJNI/d9njywszP33pndPXPn3HN+R0gpUSgUCkXzRWvsASgUCoWibihDrlAoFM0cZcgVCoWimaMMuUKhUDRzlCFXKBSKZo61MTpt1aqV7Nq1a2N0rVAoFM2WRYsWbZdS5lbd3iiGvGvXrixcuLAxulYoFIpmixBijdl25VpRKBSKZo4y5AqFQtHMUYZcoVAomjnKkCsUCkUzRxlyhaIBKdhWyO/fLmXr2m2NPRTFLkSjRK0oFC0NXdd56qqX+PSledgdNgK+APscuRe3vXktTrejsYenaOaoGblC0QC89+Qc5r7yNQFvgNLCMvzeAIu/+J2nrpza2ENT7AIoQ65QNADvTpqDr8wXsc3vDTBv+vcE/IFGGpViV0EZcoWiASjJLzXdrus6fo+/gUej2NVQhlyhaAD6H9oHIUTU9jZdcnGnuxthRIpdCWXIFYoG4JKHz8GV5sRqswCgWTQcbgfXPDfG1MArFDVBRa0oWhzLF69iyZd/kJadysGnHEBqZkq999mpdwem/P4Ysx6bzd8/LaPzHh05bfwJdOvXud77Vuz6iMao2Tlw4ECpRLMUDY2u6zx8wdN8/+4CQoEgVrsVIQT3fnQLex3Sp7GHp1BUixBikZRyYNXtyrWiaDF8/+4CfnhvAb4yH8FACG+pD0+JlztPeYRQMNTYw1Moao0y5IoWw6cvf4W31Be1PeQPsXT+skYYkUKRHJLiIxdCZAIvAv0ACVwkpZyfjLYVCjN+++YvXr9rFhuWb6L73t244K4z6LF3t7jHSF03f0MYbpddjWAgyKK5v1G0o4Q9D9mDtl1bN/aQFPVEshY7JwGfSilPFULYARVPpag3fvzgF+4/+wl8ZUb89Y6Nefw6708e+fIO9hjUM+ZxR513GH9+/0/UrFzTNPoM7lXxuiivmA+e+pSfP1lCqw7ZnHLNCPoN2aN+TqaeWP3nWm4YehcBbwApJaFgiOMuPYqxEy9QUTK7IHV2rQgh0oFDgKkAUkq/lLKgru0qFGZIKXn6mpcqjLixDXxlPp4f/1rcYw89fTD7HtUfZ4oDBNidNhxuB7e9eS02uw2Aoh3FXDpgPDMefI9/Fiznh/cWcPOwe/nkpS/r9bySiZSSCcc9QOG2IsqKPXhKvPi9Aea8+CXzP1RBBrsiyZiR7wZsA14WQvQHFgFXSykjUtmEEGOAMQCdO6uQK0Xt8Hn8bF+fZ/reiiWr4x5rsVi4453x/PXDPyz64nfSslI5/KwhZLXOqNjn7YmzKdxWTMBnpM0bNwk/z1zzCkNHHYzdaU/eydQTK5aspjivJGq7t9TH7OfmcuCJ+zXCqBT1STIMuRXYB7hSSrlACDEJuBm4vfJOUsoXgBfACD9MQr+KFojdacPhsuMp8Ua9l9Vmp0GWUvLHd3/zz4LltOqQzUEj98fhciCEoN+QPWK6ShZ8vLjCiFdGE4LVf66j98DuyTuZesLn8SM0c/eJ2XVTNH+SYcjXA+ullAvCr9/GMOQKRdLRNI0TrxjGe0/OiXCvONwOzrplJAB+X4Bbh9/Hv7+sIOALYnfaePqal3n823vovHuHuO1nVpqdVyYYCJKek5q8E6lHesW42Tjcdo44a0gDj0bRENTZRy6l3AysE0L0Dm8aCiyta7sKRTnbN+zg/cmf8PbE2axfvokL7jmTY0cfid1lx5nqxJnqZNStIxl+8VAAZj36AX/9+C/eUh+hYAhPiZfiHcXce8bEavs65drjovTBLVYL3Qd0pV23NvVyfsnG7rAx/qVxOFx2LGFJAGeqk257dmHYRYc38ugU9UFSMjuFEAMwwg/twCrgQillfqz9VWanIlHmvvY1k8a+ABjuEqFpjLplJGdPOBVPqZf8zQW06pBd4btev2wjF/e9Fj0UHU5oc9p4bflkWnXIIeAP8P27P/Pb13/RunMrjrnwcHLaZQEw89EPee2Ot7DarQQDQbr06cS9s28mq01mg513Mli/fBOfTv2SvM0FDDp2Hw4auT9Wm1LlaM7EyuxUKfqKJkv+1kLO6XoZfm+kz9rhsjPpx/vo3r9r1DGX7XtjzEVPu9PGy/9MIjUrlWuGTGDz6q14SrzYnTY0i8YDn9xW4TsvLSpjxZLVZLXJrNYdo1A0FCpFX9HseGfibAK+YNT2gD/I12/9ELU9f0sBa5aui9lem665tO6cy9sTZ7Nh+aaKhT+/N4C31Mf9Z0+ifGKTku6m/6F9lRFXNAuUIVfEpKSglBVLVlOUV9zgfW9Zs413nvgY0ydGKZF69HZdl4B5tIYQglunXwPAVzN+iJrlAxTtKGHDis11GbZC0Sgoh5kiCl3Xee76V/n4+c8NP7E/yNCzD+bqZ8dgsVoaZAwfPvOZuREHLHYrh55+YNT2nHZZtO/ehjVL10dsF5pg2IVH0GOAkcJvc5h/7aWuY4/xnkLRlFEzckUUsx79kDlTvsTvDVBW5DFqS874nlduf7PBxrDu3w2EAuaKhPsPG0DPfXYzfe+2GdeQmpmCxbrzqy2l5Mtp3/Lpy/PQdZ29DulTEc1RjtAEHXu3p3Xn3OSdRA3YFbVeFA2HMuSKKN55/KOoQsG+Mj8fPP1pzFlysuk3ZA8crugsSqvdyphHzot5XLc9u3D7rOsQWqWvtjT84E9ePoVrD76dz1792vQm0Wvf3fCUNmzCzGevfMWZHcdwjPUMzup0KZ+98lWD9q/YNVCGXBFFcYxCwd5Sr2lYX31w7OihuDPcEa4ch9vOkJH7075725jHSSl588H3CfqjF0l1XbJs0Uq8JtmNUpd8Oe17bhh6V4PdrOa+9jWTr5jKjo1GpO72DXlMvuJFPn/9mwbpX7HroAy5IopYmYGd+3Ssdx+5lJJ3n/iIi/a4muIdxdidNtzpLtp2zeW8O8/g5tevinv8rMdm8/s35vloekgn6I9dQCLgC7B26Xp++/qvOp1Dorxy+5umTz6vTGg4F5Zi10Ct7CiiuPzxCxh/+J34vX50XSKEwO6yceXk0QkdX5RXzPfv/oy31Mv+w/emY6/2Mffd/N9Wfp33J6lZKew3bAB3nfIov3z6a8X7wUAIi9XCWY+ezLGjj4zbbzAQ5PW7Z8as9pPITDvoD7LqtzUMOLxftfvWBCklekiPuBHGEv/atmFHUvtW7PooQ66Iovd+PZi84AGm3/8uK5aspmu/Toy65eRqCzcA/PzJEu4+7TGEgFBQZ+qt0znpimFc8tC5gCETO/2Bd/nhvZ/xlHgpKSjFZreiaRoS8JR4otoMBUNMufF1jrnwcCyW2E8E37+7AG9JdAWgco4851C+e2d+hEZLVawOG227Ja8AQygY4tU73+L9yZ/gLfHSsXd7rpg8mn2G7kmbLq3Y/N+2qGPadGmcBVdF80VldiqShqfUy+ltR0cVbnC6Hdzw6hWs/n0N7zz+EX6vn1CwZr52i1VjxvoXIiRnKxMKhTij/RgKtxWZvp/TPovM3HS2rN2Oz+MnFAhF+fs1i0arDtm8tuKpWrmQAv4AW9du599fVlCSX8aeB+/Oh8/O5fPXvolwoThcdh77+i42rtzCY6OfqSL+ZWf81Ms57IyDaty/YtcnVmanmpErksbiz39Hs0Qvu3g9Pu4763GQEj1Uu4lDKKhz/WH/Y69D+pDZOoPDzxpClz06AvDXj//yxGUvxDTiAIXbiioWFa12C063A5/PjyYEoaCO0AT9D+3L+Jcur7ERl1Ly1sPv88Y971QYbM2qYbFoxg2jSvKS3+tn2n3vcPf7N6Fpgqm3TWfrmm207pLLxfeNMo2RVyjioQy5IibF+SV89Nxcfv3qL9p1b8PIq46tMJ5m6CHdqNhaFQl6DWfgZqz7ZyPr/tmIxarx9mOzGf3QOfQZ3Iubjr4natGwKsFK4YZBf4ig33DhhAC7y87FD4zi5KtG1GpcHz0/l2mVjDgY5xvrnKWENX8ZUgK99+/BrdOupmu/zlGqiwpFoijXisKUvM35XLbvTZQUlOL3+NEsGjaHjTvevp79hu1tekxpYSlntB+DzxPbB51M7E4bu/Xvyj8Llte5rbTsVGZteTGuDz4WozqPZdv6xBcojdl/H9b+vYH8rYVYrBYsVo2LHzibkVceW+P+FS0HJZqlqBFv3PM2hduL8IeNsh7S8ZX5eGz0szGzEFMyUhg3+aII90p9F/pNhhEHo+anWXm0RMjfUhj3/arVeqx2K79+/Rd5mwuQuiToD+Ir8/PizdNYOPe3Wo1B0bJRhlxhyk8fLTLNfiwpKGXr2u2mx0gp+eTFeRGGvFZPfCKxG0AwEJ30U1s0q4XUzJRaHdu1X6eY79ldNgYfP5DsdllY7Vb6DO5FZqt0UxeU3+Nn1mMf1moMipaN8pG3YKSUrPp9DflbCuk1cDfSs9Mq3kvJSGHbumh3gR7ScaU6TdtbtmgVq/9YY5pVWbOBgTR1tlcdS2I3CZvDaiqHW47dZefka0bUuujC2MfO57YR90e5lGxOGz0GdOO2GddUFL7wlHo5KTO2xEDeppj1WBSKmChD3kLZvmEHtwy/n82rt2CxWgj4Apx1y0jOuf00AE6++lieufplvJUW8Kw2C3se3IeMVummbW5YtjFm0d9E0CxaQhIAQoDNYcOd7qZga3y3ht1po7qHgpOvHsH5d55ek6FG0P+wvjz0+f945fY3WfX7GlypTrrv3Y1jzj+MQcftE+F3t9mtaBYLesj8xrLfsAG1Hoei5aIMeQvlfyc9zNq/10cYzun3v8u29Xkcf9nRDLvoCFb8uppPXpyH3WklGNDpvHsHbpkWO0W+S99OBEx0vhMlUR0XKQ1XSO/9uvPrV39FR6wIo85mqw7ZtNutDX/+8I9pOxarhT0P2QM9pDNmr+vJbJ3Badcfz6AR+9Z47H0P7M0jX95R7X5Wm5Who4bw+RvfRkW1uFKdnDb+xBr3rVCoqJUWyMaVmxmz1/Wm0SVGOr6dQcfuw60zriZ/SyErl6xGs1pY89c6/N4ABxy3L7vt1SXq2IA/wHEp55ga5ERn2zXB7rLTe78eLFu4Ek0T+H0BuvbpyKjbTuGgk/bHYrVw2cAbWbHYvPRbbqccgoEQJXklBMLuIGeKg7MnnMKZN41M6lgr4yn1cvepj/LrV38hdZ1QSKdrn048OHcCOe2y661fRfNHJQQpKijOLzX0uKOz4ZFS4ivz8fMni5n7ytcMv3gov371J4+PeR6pS/RQiOn3vcNxlx3N2EfPjzh227od2BzWmCnwyTbmfo+fFYtX8daG59m+IY+c9tlRC5ZbVm81PdZi1Tjg+H35dOq8CiMO4C318fpdb3P8ZceQku5O2lgr40px8sAnE9iwYhObVm2lS5+O5HbMqZe+FC0DFbXSAum2Z2fzxJ1KeEt9zJnyBUU7inn8kufwe/wEfAFCQR2fx89Hz33O0vn/RhyTkZselcVYTo99uuFOd5m+V5cIRU+Jlxdvnk5W28woI776z7WUFJSZHiel5N8FK0wXQW0OK79/s5T5sxfy5w//1FvRhw492jHw6P7KiCvqjHKtNFGklKz+Yy2hYIjd+nepVaJKPGY99iFTb50eswoPQKfd25Oek8bS+cuiamQKIThh3DFc8eTFEdsfHf2MUROzkttGs2jkdsxhy5pogahk0bZba178cyIOl4NgIMhrd81k5sMfxNV00SzCNPLFYrMYLianDSklaVmpPDT39rgqjgD5Wwv5dtZ8PMUeAoEgn06dx/YNebTr3oYxD53LgSfuV+fzVLRsYrlWlCFvgixfvIo7Rj5McV4JQggcKQ5uf+s69jqkT1Laf/2eWbz5wHvouiGtaubusNmt6NJIMzf7ighN0Gn3DuRvLsBqt3LY6YPxlPr48YNf8JX68Hn91c76k4nFpjFk5CDadm3Nknl/smzhylq1Ux51U/nGJYSgTddWjJ86jg3LN9G1Xyf2OKBXRKz7/NkLue/MxwFjraDqDcLhtjPhzes44LiaL6QqFOUoQ95M8JR6OavjpZQWRroEnClOXl/1FJm55up/ibJ88SquPfh204XO8nhrV6qTgC8QoU8ShQBNq+TzLrdpDf91qiDWDDsZlM/QhRAgoNteXXjoswm4Ul0xVR+r0rVvJ6b8MbFexqdoGagU/WbCD+/9bDpD9nv9fPTcXHye+MaiOuZN/w6/LzpE0Jni4KCT9ueUa0dw+aQL48aDW20WLFUXLiWNasQh8QSh2iClxOfx4y3z4S31sWLxaqbeOh2AX+f9aar6WJWNKzdHvN6+YQefTP2SL6d9R2mRuS9foUgEZcibGAVbCyOiKMrRQzqv3jGTkdkX8MA5k0wLMCRCKJZCoYB9jx7ApY+ez98/LYuZCalZNYLBUI31xHc1Ar4AX7z+LZC4DEHlghVvPfIB5/e8kmeueZlJl73AmR3G8Mtnv9bHUBUtABV+2MTY69A+xAviCPiCfPfOAgq3FfHgZ7fHbWvFr6uZ+ciHbFi2ka57dsbutLNh+SYsNktUGr0e1Bk0Yh8+f+0b5k3/PmabyZCjbeqUu1F8Hr/hMophp8uv4d5D96w2rNLhsnPRfaMAw731+p0z8VdJnrr71EeZuWkKrlTz6B6FIhbKkDcxeu3bndzOrdi4YnPMfQK+AH98/w+bVm2h3W5tTPf5+ZMl3H3qo/h9AaQuWbZo1c43BRXCVBaLhmbRGDvxAjJz03l30sfV+nqbFFV99XVEs2h06duRwccPZMWS1WS1zmDV72tZsWR1xMxbs2jsP2IfAv4A0+97tyKGstwlZbFa0Cwafo+fNl1yueShczjopP0B+Py1b0zdW0LT+HnOElVYQlFjlCFvJPy+AG8/9iGfvvQVoWCIw886iFG3noI7zcWIMUfy0m0z4oYG2uxWtqzZZmrIpZQ8cenzsXXBw/ao18DdyMhNZ+mP/zL5iheZft87pgamqWF32gzXjoC9j9iTYy44nEcufCpqhlsb9JDOf3+uY+PyzZx9+6mcdfNI1i/byFUH3obf68dX5seZ4sCV5uKyiRdwz+kTWfT57xXhllKX2J027v7gJvY9qj9SyiglR7/XHxXOCcbnVpPr7yn18uP7v1C4vYj+h/Wle/+udTp3RfNFGfJGQErJrcfexz8/La8wtu8+MYcFHy/m2UUPc/T5h/HGPW/jiWPI/d4AXfqYV+sp2FZEwbb4YlIAK39bgxBU+MO3b8jDYtXqJZ0+mYRCIc68eSSbVm0hu20Wj178DHH9UTVE6sbC5ht3z+LwMw+iY6/2vLp8Mp+/9g2r/1hLr313Y+g5h5C3uSDCiFfmrx//Zd+j+pvK8R5y6mC+nPZd1JOPHgwx8JgBCY1x2aKV3Hjk3ei6TtAfRLNoHHTSIG567Qo0TS19tTSUIW8E/vrxX/79eUXEjDngC7Dlv23M/3AhB59yAA9//j/uP3sS2zfkRQlROdwOjjrvULLaZJq270xJrGSYmdxs+SKmzW41XXRtCoQCOtPvf9d0VptUhGDBx4s5cdww0rJSOfnqyFJwS778A6vNGmXI/d5A3IIXew/dk4NGDuKH9xbgK/MhNA2r3cqYh8+JWVy6Mrquc8dJD0eFqP74wc98NeMHhp59cA1OUrErkLRbtxDCIoRYIoT4KFlt7qos+2UloWD0bNtT4mXpT8sA2H3/nry6bDI3vHQ57nQXmtX4qDSLxnGXHsWVT10cdXw5rhQnB40chM1hq/UYgybja0okYsStdmvdZHU1EfcadujZjlAw+mZntVvptme0qFg5QghuevUK7p19CyeMG8Zp40/g6Z8f5MRxwxMa18pf/4sy4hCWVXjxi4TaUOxaJHNGfjXwN2AuVq2oILdTDlZ7dLEDh9tOm865bF27DVeai4A/yKMXPRPh+9VDOh88/SkjrzqWNl1yY/Zx3QtGUtFvX/1JqBbhgvU+220A6lrgwlvq44Xxr/Hn938z+sGzyW6bRWlhKZpFw5XqomvfTuy+f0+Wzl9GoJJv2+awcuIVw+K2LYSg/2F96X9Y3xqPKxQMxaygFG9dRbHrkpTMTiFER+BV4D7gOinlcfH2b+mZnQF/gHO6Xk7+1sIIg+lwO3CnuygtKEUP6QhNizAQlUnNSuHV5ZMjqvpURUrJjAff44PJn1CwrRApjdmiEUIoW3wseKJYrBbSslPJbpvJmr/XI4ABh/fjhlfG4Up18vTVLzNv+ncEAyF67duda54bQ4+9u9XbeELBEKe3u4SiHcUR251uB5c+dj7HXXpUvfWtaFzqO7PzCeBGQFmGBLDZbTzx/b30Gtgdm92KzWGj3W5tCAVD5G8uwO810uNjGXEAT7GHD5/+NG4/L0+YwfT73iVvcwF6SCJ1iRBw02tXqgUxjOSmRGqDhoIhCrYWsur3NYQCIYKBEEvm/cl1h96Bw+1g/NTL+ah0GnM807ntzWtYOn8Zn740j6K84mrbrg0Wq4Xb3rwWh9uB3Wm4fpypTnYf1JNhFx1eL30qmjZ1npELIY4DjpVSXi6EOAwYbzYjF0KMAcYAdO7ced81a9bUqd/mxKZVW/jqze/xewMMPn4gvffrUfFewbZCQkGd565/ha/f/LFG7e5xQE+e/PF+0/dKi8o4ve0l+L0xQhDjJLq0BJypTh789DYeG/0cW/7bWqvQRWeKgwNP2h+bzcI+R/Vn7d/rmfXohyAEmiaQumTCW/UnlJW3OZ8vp31H/pZC9j6iH/se3V/doHdx6k00SwjxAHAuEAScGD7yd6WU58Q6piW5Vj556UueumIqesioBGN32jjmgsO5YvLFEbPBs7tdxtY15tXpzRACDj7lAG6feb3p+8sWreSGoXdRVpRAKr9Gi3uWsrvshIIhXClOdturM8UFpaDDuuUbCcYp1FyV8lBNh8tuxIdX+Tk53HZmbnoRd5rK1lTUnXpzrUgpb5FSdpRSdgXOBObFM+IticLtRTx1xdQKV4nUJb4yP3Nf/Zo/v99ZR3LHpvwa17o0Kr/HXorI7dQqbuX4CFqYEQejulAoEKKkoJS/fvyXDcs3s/bfDeg1jNYpj7f3eaKNOBiG/uc5i5MxZIUiJuo5rB75+ZMlaNboghC+Mj9fz/wBgN+++YsLel0ZtXAVD6vNwhWTL6bvgb0rtm1bv4O7T3uU49PO4cTM85hy4+sMOm6fhFT5WjqhoF5h2E0VFIWxuGyx1aK4hyS+HLBCkQSSmhAkpfwa+DqZbTZnLBbNPOEwXOVd13XuHzUppraJpgmkjFbX03XJhmUbKSkoZcPyTUy9dRpL5v0Z4fP+/LVvcKe7mnSGZlOmPP7c7rRz4IkDGTvxAl65/U2+mvF9jbRogoEgHXu1Y83f6+nUu73yYSvqBVVYoh4pKSjlzA5jojRPHG47j319Nw6XnSsPuBVvqdf0eFe6C08MH7fVZsXhthPwBZKiMaKIJC07lZFXHUv/w/rSb8juaJqGp8TDaW1Gx9SwsVg1NE0jEAgaolmaRmqmm7JiD0IIUjNTuG3GNfQbskcDn41iVyGWj1yl6NcjqZkp3PT6VTx47pNoQlQU8T39hhPpPbA765dvQsrYM+ZYRhyMmV6wsGmm0Dd3hIDM3HTeeeIjXr97Frkdczj5mhH89vVfBALm11xogqHnHMLwi49g/ocLsTqszH5mLgVbCyt8595SH7cMv5/XVkyOKa+gUNQGZcjrESklrTvlcNlj57FjcwHp2WkMGrEP7bu3BaBDj7bkdmrF+n83NvJIFZURmsam1VsrMkO3rt3Oc9e9ihAiZhGJfYbuyQ0vjQOg30F78N07PxEMBKMWQPVQiM9f/4bTx59Yr+egaFkoQ15PlBSUcvMx97Bm6XqEJggFQvQ5sDcjKmXdCSE49drjeGLsC404UkUEYX1zs/T+WEbc4XZw5s0jyd9ayFsPv8+CjxcTCgRNI5H83gDb1uclfdiKlo0y5Elg3b8b+PSleRRuK2LQiH058MT9mHTZC6z8bU2EQfh13p+c1+MKnvzxPlp3agVAwdaixhq2ogp2p40Djh/It2/PT2h/oQlsdivn3nEau+3VhUv2up6iHcVxNV6cqU4G1EJfRaGIhzLkdeSrt37gsYueIRgIEQqG+Obtn+gxoCv/LFhuGna2Y0Me1x3yP15dMRmLxULrzq1wpjrxlpgveCoaCAEnXTmcfY/uz7ezEjPku+3VhYe/+B/p2Wm8eudblOSVxDXidpedzrt3qLdMT0XLRcVC1QFvmY+Jo5/F5/FXyNJ6S7wsX7zKKHIcg/ytBSz87DcADj71gAq9DEUjImHR579z23BzyYOqOFMcnHfn6RWiZYs//920uo/daaNDz7Z07deJ8+44jYnf3IXFJLdAoagLypDXgaXzl5km3PjK/HHjhf2eAPecPpE/v/8bp9vBoacekJB4k6J+WfnrfwnrsKdkptBvyO4Vr3M75Zh+hkII7v7gZqb8PpEzbjwJhyuxoh8KRU1okYbcU+KJLSZVAxwue8wFMLPCEZXxlfm4dcT9rP17HZ+98nXMdhSNj81hpaqNLtxWxEPnTa54fcq1x2N3VXmyEsZCqKee3GbSOxd9xyj0bcPRix9G6moRtaXSogz5iiWrGbvPDYzMvpATM8/nzpMfqVFqfFV2H9QDV6qz9gOS8PbjH6s0+iZMbqccMltnRIURBv1BlnzxR4VU7R6DenLtC2NJyXDvnJlLKMor5vrD/scXb3yT1HHpJZORBTdAYCGEVkLpq8jtJyL16mu1KnY9WowFyducz/WH3cHKX/8jFAwR9AdZMGcRNxx5V61nwxaLhfs+upW07FTcaS7szpqtHeshHaQRt6xomuRvKYxZu1SzahHqkkNHHczIq46N1GSRhqtt8rippj702iD1Qih5AaicMBYAvQBZNi0pfSiaFy3Ggnz8whdRP8igP8SmlVv4O1wnMx5SSj5/7Rsu3+8mLuh9FS/e/AZFecX02Lsbb218gZvfuIpjLjyiRr7uUDDE0HMORqtDXUlF/RL0B/GWeE1rf7rT3bTu3Cpi208fLTSPXBGw6rf/kjOowF8g7CZv+MD3bXL6qAek1JHB9Ui9oLGHssvRYgz52r/Xx6y4s2nV1mqPn3zlVJ4cN4Xli1axYfkm3nniY8YNvBlPiYd/f1nJc9e/yuxn58ad3VeNTpFScuvw+zhx3DCsdhUJ2lTxlvrC1ZUMY65pAofbzrXPXxq1qJ0Wo/ReKBAiNSs1OQPSckGaPSUIsLRPTh9JRnq/Qm4bgtx+LHLrQeh5o5VBTyItxpD3GdwLhzs6YkAP6ezWP3bFczAkYj+dOi9C9S7oD5K/tZBJl0/hhqF3snHF5mrHUFUfPBTU8XsDvPnQe9iUIW9wzGbZ8bBYNTr1bs8R5xzCpB/uY/DxUdpFnHz1CJwpkd8zzaLRafcOdOzZrk7jLUfYeoJ1N6LTQByIlAuS0kcykYF/kAXXgL4d8AIB8M9H5o9t5JHtOrQYQ370+YfhTndFLCzaXXb6H9aXbv06xz32n59XYHNEG1pfmY+v3/yRoD+xkLVYs3U9JOstskGxE4vVQrvubXCnu+nYq12NJWWDgRC99+vBTa9cQff+XU33OeC4fTnjxhOxO224M9w4Uxx07NWeuz+4KQlnsBOR9SLY+gMOECkg0iDjPoRtr1q1JwPLkZ73kf5FSY+gkmUvA1WlfwMQWIoMrkxqXy2VFjMNTMlI4ZlfHmTKzdP4+ePF2F02jh19JGfdenK1x+a0z4qodh+Bcm83C4QmOPu2kzn3jtMBePeJj5l667SYYaJmAlmaRSMtJ5VfPvuVD5/+lOL8Eg4+5QCOveRIXCk7o5fOuf00Thg3jH9/XkFWm0y6D+ia9DwBYclB5MxAhjaBXgjW3RCmfvP4SOlHFlwFvh9BhG9slg6Q/TpCy07OYINrMS1DJWwQ2gzW7snppwXTYgw5QKsOOdzy+lU1Pm6PQT3J6ZDF+n83Rb2nCjc0D6QuefOh9zlu7NFktcmkKK84po57jwFdWfn7mqji1Da7lWAgxN2nPlrhZluxeDWfvjSPpxY8EJHsk56dxn7D9q638ylHWNqBpfYuG1k6xTDieHeeb3A1suAmRPaUpIwR+2AI/EnUrFz6waa02ZNBi3Gt1AUhBGfceFJMn6qop6totVuw2CxY7OYp3XanLcofq4iNxWph4VxDGiGW+0DTBGv+WR/1BGZzWLnogVFRayU+j59Nq7by+WtNOFpEL0EvmYK+42z0guuQ/t92vln2FobfujJB8P+A1MuS0r9IOcdw/0TMG13gHpW8WX8TR4a2oRc/gZ53EXrxI8aTVBJRhjxB/vsr+scNhnHIbJ1RL30G/SFCgRAhEx+8EAJdyhqVHWvxCKN0G8BXM34w3UXXJQFvdERIhx7taNu1dcy1kp8+apoVr6RehNxxApRMhsAv4P0YmXcuetm74R1ifX90pO9LpKz72o3QshGtPgDXKaC1BWtvRPodiLSb69x2fSD1YqT3MyPSJub1qUF7wdXI7cOg9EXwfw+lrxjRO4GlSRitgTLkCZLbMdtU3Epo4Clu+IVKKSVBn6oQVBN8ZX72OtR4lM/blF+jY9cv30R6dqrpTF7TBFltM5MxxKQjS1+F0DZ2zrql8f/iewwj5TwCMHvik1B0B3LrgUhfYmqQ8RCWNmgZ96C1/hat1WyE++QmqS+kl71nnHPhLcjC65FbD6jz+cuie0GWAOWyIAGQpciiO+o83nKUIU+QoWcfbKpaF/AGaz0r1qxN74u8KyM0waSxht+3S9+ONTo2FAjx8ydLSM9OizJANqeNEy47JmnjTCq+L4mOGAEQEPgHkXodaK0AV5X3pWF8ZAmy4DKkXlIvw5P+n9HzLkbfdgx64QRkcH299JPQWIL/QdEdgK/i3JGldT9//wKiFlwAAr8jTfMBao4y5AmSmZvBZU9ckNQoFT2ohLIaklAgxIKPF1OwrZAxD59XI/lgKSXvPP4x104ZS/sebXCmOnCnu3CmOLjyqdH03Ge3ehx5HdCyzLfLIGgZCEsuotUnkHYDWHpg/gUX4Ps66UPTyz5E5o0G/3cQWg2ed5A7TkQG1ya9r0SQng+AGIbVN6/2DYuqN8lybJg/DdWcFhW1Ulc+efFL0xurovkQDAR545630TQNPVZIaQyklKz5ax0v//MkK3/7j7IiD70GdsdpkmjWVBAp5yMDi0FW1mWxgLU7wtrV2EdLRaScgx5cDZ4V0Y1IHWRpUsclZQiK7yVyoTVkzIBLnkRkPprU/hIbVAmmhlzqIOuw8Os+A0pfJfLJyA6uk5LmXmqRhlxKSd7mApwpDlLS3Qkft2yhSl5oalislmolg6sy+7m5RihhnGo+pn1ZNBwuO0IIegzoVqNjGwvhOAyZchmUPG3EbRMCSydE1rPR+zqHIj1vEynGBaCD4+CkjUnqJUj/72C6kKqHXRENj3AcgfTMMjHaEuxDat9u6lXI4CrwfWd8BjII9n0QabfWbcCVaHGGfPEXv/PY6GfJ31oIumTgsAHc8PI40hLQwUjJSKmT7K0iuQhNsFv/Lqz89b+IeH7NqqEHY8f3C4yFz+rarhqlJKVkyMmD6jTmxkBLHYt0n2XEcmvZYN3dfCZoHwyOQ8H/bdiYCcAJKRcjkqDhImXQWPjzvAPCQnTYYxhhLCo3+GKo/QCwH2Jy/hcirDVbU6mMEHZE1jPI4BoILgdrV4S1R9KGDSAao6DBwIED5cKFDR+utebv9Yzb72Z8ZTsfcax2K70HdueJ7++t9vg37n2b1+6YqYpANBGy22Uy+acHuO7Q/1G0oxg9pCOlJDUzhbxNBTGP0yxatYlcGa3S8JR4KyRp9ZDOLW9czUEn7Z/MU2hySKmD7xuk9yMQToTrZIQ9OTVG9aJHoOx1YhrwCuxgaYvIegFhbdi1h53nPxuEI3z++zXoGOIhhFgkpYwS+WlRM/L3Jn0cpYAY9AdZ8etq1ixdR5c+neIef9YtI/nh3Z9Z8evqmPtoFoGUxiO/M8WBt9QX9Qhvs1uxp9gpzU9OwkVLJRTUKS0s49Xlk1ny5Z9sXbONeW9+z29f/RX3OD2kY3faYmZ2AnTfuxu3zbiGhZ/+imbR2G/43jVyw9UFKSX4PkWWzQQZRLhOAtcJCGG+OCv1fAiuAUtHhKWV6T6JIoQGzsMRzsPr1I6UIfDPh9AGsO0J1t7gmUb1RhzAD6F1yLzzIPcbhGi4GqfJOv+GpkUZ8vXLNpnOxKw2K1vWbK/WkFssFjRb7EAfd7qL48YezZk3nUjQH8LmsHHnyIf5/du/kUgsFg2L1cJlT1yIHgrx7HWvqljwOFisGkLTCAaCWKwaoaAesdhcuK2Im4+5l+lrn2Xg0f3ZsmYbj1/6fPUNC0hvlUZxXmnE01k5DpedcyacSnp2GkeMSp5vOFFk0S3g+YRyX7UM/A7e2ZD1kmFoyveTIWTR3eB519Anl36k8xhExv210l1JFjK0GZk3CvR8kOH1C/u+MXziYERuSCL1WKSxwOr/CRwH1e+AdwFalCHf69A+/P3TsqiZWMAXqFbKtpw1f62L+d5d79/EgMP6AkZd0LF738D2DXkVrhihCYaNHoor1ckTY55XRhzDPeL3BijJ3xkVYbVbyW6XSf9D+/LPzytwuB1YLIJlC1chq4QNeUu8/Pn9P/Q/tC/zZyforpPG36Pz7mD+hwv5/ZulLFu8Cj0YIqttJuMmXcSeBzeOBogMLAPPHCJnrh4I/Ar+HyIWHWXpFPC8hxH3HL4heecitRxE+i3GPtIHvm8MYS37IIQ1vtJnUs6hYDyENgGVFqH9i0CkgiyKPkCkgzRJ0JIYN4NmiAxtQXpmg56PcAwBe/0WWG9RhvyEy4/hw6c/JRgIVczMHW4HR517CK3aJ6b5kNMum40rzbXH7z7tUWZunILVZuXTl+axY1N+xE0j6A8x54UvmPPCFzWOmNhVyd9cyAdFr/HOxNl8MnUeQX+QQ04djN/rY96MHyqSrTQtWo0QAEFFuTWL1YIQRNXXNKNj7/bsvn9Pdt+/J2CEJXpLfZE1N6vhj+/+5u2Js9mxMY/9hu/NyKuOJT1GYYmE8f+EqVKgLEP6fkRUjh4pe5VoV4UXPG8i026G4F/IvAuBoBFCh450n4lIu7XejIrUCyGwhAgjXj4uUgBnlTE7wH2mcS5R0SIBsEdrvpv2K0Pg+xrp+wEsrRGukxCWtrU9jTohfd8i86/A+Bz9SM80sO8Pmc8gRP2Y3F02IUhKGRWWlpmbwbOLH+HIcw4hs00GHXu1Y8wj53Ll06Ojjl/z9zpeum06r901i20bdlRsP+d/p8YslhwKhPjz+38AWDBniWlkhB7SlRGvhJSSm4++l8EnDOSUa4/j7AmnMvScg/nije8iMmZjxXwH/UH6DdkdgINO2g+LtfofisNt57ywnG05VpuV1MyUhA3cnBe/4Jbh9/HjB7/w7y8rmfnQB1zaf3zdo5q0zHCYYFXsRsRJZfQYfUmvIU+bPwZkYTgG3AP4oGxmrZN7pJRI34/IkueRng/NdVhkkJhZc0JDZL8EtkHG+aAZf2WvAukYRr4cF7jPQ1jaGi4kGTvEVEo/Mu9sZOH14HkDSp5CbjvGMOoNjCELfC3GzSr8+5dl4FsA3o/rrd9dbkYupeTdSR8z/f53Kd5RTG7nVox5+DwOPW0wALkdc7jh5XExjw+FQtxx0sMs+HhxxbbX75rJSVcNZ9wTF3HUuYfy1sMfxHSxlM/Acztko2kiygAp2dto/l7wL5fvexMWmxVNE4SCoZhKk+W+ciEEdpeNi+4fVRE6mt02i+teHMvE0c8SCukV4YM9992NDcs24Sn10rZra8ZNurBOrhOfx8ez170a4V/3+wIUbi/m7Ymzuei+UbVuG8eRwJ0mb2gI1wmRm2z9DSGsqlh7IIJ/IWXVeHAAD9LzVsKLeVJ6wPetkaJeNh1CKw03jnBA0X2QMyMiskRYcpCWTsZ+kYMFxzCEfSBSaOz0iXvC6x4hcJ4Mob9BpCLcZyOtfdHzxhiZn4C0H4jIuNeQ7q08xrK3ILCUnTN9w4DKguug9Y8NulhK4DfMswY9SM97CNeJ9dJtnQ25EKIT8BrQFuOTeUFKOamu7daWmY9+yBt3z6qYzW1ds51HLnwKh8vOAcfFDqPylHqZft87zH72M0oLo38AHzz1Kfsdszf7D9+bUbeezONjnovSWAkFdfY8xDAQJ14xnK/e/AGfZ+es3Cw2WRFOnEOiV1Nl3mLV2Pfo/miaRlp2KseNPZo+B/SK2Oeocw9l4NH9+eH9XwgFQww+YSCtO7UyRMYCQWz2xNPyY/Hfn+tMC2YHfAEWfLy4ToZcaG7IehlZcFk4G1MAGiJzIsLSJnLf9NuMRUXpxfjpaYAdkX6nofUda2acoKKh9P2ELBhrtCN9RGQ9yjKgDFlwHaLV+5HjynwEmXdueHbuA+EGLQeRdqUh3+pfDFT9rH3gnQUZj6G5hiOlH7YdBfpWKtw0/h+QO06H3C8jF3M9H2AeDeOH4FIjaqbquYU2g14ULsiRzPls+U3KjLp/92KRjDMIAtdLKRcLIdKARUKIz6WUydNoTJBQKMSM+9+NMrC+Mj+v3P5mTEOu6zrjD7+T//5cGzMkTeqSdx6fzf7D9+bQ0wbzxevf8Mf3/+At8WKxWrDaLJzzv1N58vIprP17A30P7M1JVx/LOxNnV/hsUzPcFOeXqll5LbHarFz+xIV06BG/kEJWm0yOu/SoiG1CiKQYcYD0nLSY7rGsNnWXNBb2/pD7PQT+AIJg28s09FDY+kDO+8jS541kH2svRMqlCFuvsPyqmUFxIZwnGNmVJU+C50NjP+exiLRrEJoxfqmXGUa8utT04HJkaBvCkmvM3r1fgr4NMp4E3+fg/xm0NpA6DqFlIgPrw9mNZkJeQSi8CWntAKGN4YXRyi4VHWQxeOeC67jKFyzG4CRVjacMbUcWXGlcL2EFLMj0u9BcI+KfZ9WWZcBYWPb/CFobhPtUI2nK1h/DbVRV0sCFcJ9aoz5qQp0NuZRyE7Ap/P9iIcTfQAegwQ15WZEnYgZcmY2rtsQ8btHc31j3z4a4ccUAJQXGl9pitXDvR7fwy6e/8tNHC0nLTqNjz3Y8Oe5FAl4/ui53pvMLKn5PhduLm6R0Z1PE7rThTHXi9/jRNA2hCW585YpqjXhD0G63NnQf0JVlC1dFrMM43A5Oufa4OEdWj+7/DYruhOAKQAPH4Yj0m2JWARLWroiMB6K3Cwcy/UEovAFjrhU0ZsbWPZHOEZB3RriP8O/FMxPpnw+tZhs3Df+3JKYQFwKkUWA571yjH+nHeEIIu09C/0H+EvSUCxGpl2G6mFuBD1n6EsLau4o+TBhZhgyuihiZcJ+BLPoren+RZcSvVz48/xII/hseZ/hmUngL0toFYeuXwPka7ia54ywI/geUAXZk6YuQ9SzCcRBkPYvMvzi86h4ENHAdC46jE2q/NiTVRy6E6ArsDUSJJQghxgBjADp3rp8QqJQMN65UJ8V50ZKTnXrHTjFetnAVXpN44ggEHHXeIRUvNU1j0LH7MOjYfZBScn7PK01jkqtOiipHXiSSYdhS0UM6z//6KEXbi/GW+ei5T7eYM+pQMETRjmLSslOx2hpm2eeu927k9hMe5L+/1mG1WQj6Q5x/1+l1Ku+ml70HRTcT8aXxzUFu/xFyP0NUUjKU0mcUifAvAktnhOtUhCUnoj3NdQzStgfS8w7oeQjHYeA4zKhgH/qPnfrYAAHQtxgqf85jwu6XRNyADtBykXlnGQur5mcGeKH0JXCNhNTxUPwQ5vK6Eny/In3fYF7n042wRRpnnMeD91vwfYpx87GCsCOyno2YOMnAMgiuIloYy48sfSVhoS5Z+nq4nao++euh9Q8I+z6Q+x34vgC9AOyDo8ecZJL2rRdCpALvANdIGR0sKqV8AXgBjBT9ZPVbGU3TOP+u05ly07QIo+pw2bn4/p1+y7JiD79+9Sc2u5X+h/ejdZdWON2OuJXsO/Roy/CLh5q+V1pYxrZ122s8Xl1XRtzhtptG99icdjYs30S/IbtjsZgvVkkpmfXYbKbd+zZBfxCLzcIZN57EqFvrv2hBVptMnlrwIOuXbSR/SyHdB3TFnRZLrrR69NCOaCNejixGls1ApF5uREV43oGiRzAMSRBwIEufg+zXEFX8wcLaGZF2bWRzgaXhWXPVfkqRBdcjrU+B+9ydyTzx0Fojtx8NoUR0xCX4vkZLuQBdaw2FV5qcrwZyM+azdgtoueA4InKz/3vwfxV22YQMW575OMK2e+R++jbDnRJ1iXXDlZMo3o8x98l7IbgMbHsgtFRwnZR4m3UkKYZcGA68d4BpUsp3k9FmbTlx3HCcKU5ev2sWOzbm0bF3B8Y8ci77HLkXAF9O/47HL3muQkNDCMHtM6/D7rTjLfVFzJiFJmjfvQ0jxhzFiVcMx+4wnxHaXXaEphEdO1sNu+i6pzPVia/KtTTDarfGFK/yFHu4/YSHcKc5eWju7aZZtx+/8Dmv3Tlz503bG2DGA+9hd9k47boTovavDzr2ak/HXnUXlCJ/LLG/ECHwL0L3LYb8izAe5ytjJATJghsRuZ9U25XUsojt3vAbrofi+wwBLd83mM+cAQTom4ic2cdDA2GEGGquo9GDl0PplCrHx5nciExEzlsR6wUytDUcs13JsEqgYBwy9zuEVimu39YnxkKvHRw1UDcUMWSLpYz9Xj1T5zhyYUx9pgJ/Sykn1n1IdeeYCw7njdXP8InvTab8/hj7HTMAgA0rNvH4Jc/h8/gpK/JQVuShtLCMu055lAc/m0DPfXfDarOgaQJnqpMjzzmEh+b+j9OuP8HUiBduL+LvBcvxFHs47IwDY8aXtzQOPuUArLbqQ76Cgfjx9J5iDzs25nPLsPtMn16m3ftOlDvLV+bjzQffr9F4Gxs9sAKCv1WzUzHkn0m0Ea9EaB2670f0wnvQC25Cer8wRKAqIaWE0teI76fG8Df754N7NLGLHwgSN+JhnEcjpRfpeR+CK6nZ5McfXazZ+zHmCVQYi6KVEFoWWHuatBsCV+ILkcJ9lkmxCAGWtmBpHHnjZFieg4BzgSOEEL+G/45NQrtJ54vXvyFopl0tYO0/G7j2hUuxOqwITeAt8fLVjO+5ZK/rWPnbfxG7h4IhHh39DGd1HsvNw+7lzE6Xsu7fDfUSWmh32clpl0XPfZtoBZoqCE3w1YzvCSSS9JTg5SopLOWfBcujtudvKTDdv2h7cdLcVlJKZGAp0vc1MlRz91lClE6tZgcBwd8TaCgI+Zca4lTe95AF45H5YyKTaYJ/gB5bZiICGQAhiG1sE7nGVhDhjM6MRw09mG1HIovuDPu0a2DILR2ih6gXYH4zCYAsiNw3uDZ88zAZY8kkpH9hYsqmzhPBOQxwAC7j/LScKJ98Q5KMqJXvSWoBtNhsXbedsiIPnXZvH9NvGo+SgjJCgegvjh7SKSvy8NQVU/GW7JzhBQMhgoEQT105lce/vadi++t3z+LrN38g4A0QCEe6/LPApLJKLbHarew5ZA9ueGUcuR2NBaz3npzDf3+uJdDE9VmkLk1D81KzU/CV+mo1fiEEpUXREQyddu/If39GlwVr370Nmlb3OYoMbUfmXwShNYDFMELusxFpNyf3BxtcVt1ISOyuJ4l0g5RBYKGx6OYM1xQNbSLx+ZsEa3QMdo0QGYj028BxKEJLQ99xejg2vKbYEalXG6PSC4yIEUsHhOMgpGl6vwXsB0ZuCiwxdNCjLqUPPO8ivR+DpSNkvx6xsBx1SkJDZDyETLnU0JDRWoFjSEx1yoagWWR2bt+wgztPeZTVv69Bs1qwO23c8NK4uAk+Zux71F7MmfJFVJih1CX7Hr0XT46bYnrc0vmRP7QPnvq02sIENSWrbSZ2p53sthkcdvqBOFNdrFiymqw2GeRtymdHDau+NwbCIpAhc4NTVuTB7rDVypCHAiH6DO4VtX3sY+dxx0kPR4ScOlx2Ln3s/Br3YYYsuCocoldpzGVvgq0vVM2yrAu2PhD8i8RmuLGwYMQvV7nhyTJk2duIckNu7WvMtBPBdTLoG+owJoy6oOGYb937lSH+VRtSrwbH4ehF9xgyA8IezjDNjfZ7Czc4hyFsVbJ3tVbEnnOGjJtBcBWycAIi6+lqhySsu0ED66XHoskbciklNx51NxuWbw6H6gXwlni598yJPP3LQ3TZI7HKHR9P+Zznrn8tSn/FmeLgpCuG065bG5wpDjzF0YshVWsylhQmt34hQHFeCVLquNIcvHDTG1gsGprVQsBrxFFbbJYmPRsXQmC32/D7/EgTe2SxaARNnoY0i4bFqlWcW3lqvtQlQhPYnTbGTrzAVAt836P6c9+cW3l5wgzWLF1Px57tuPDesyoWtuuCDG2FwO9Eh6p5jFC1JBpykTI6XGKtDli6QGiV+Xv+79E9c9BcxyKsHZGuEeD5mNiLmAAWsA00iiLXGis4j9z5sujBWrbjMmqPlr0KZW8TofYoq95orJB+F8Jp8vnYDwCRFo43j3XTDILvK6T0I4TdWGOQXhCuJp0D0uQN+d8LlrN9fV5UvHXAF+TDZz7lysnRgldV+f3bpTx77SuRs2gB6dlp3D7zOgYcbiQCHDv6SGY/Nxd/pRme3WljxKVHRrSX0y6LHRtrPkO2OW0gpalBLndH/PeH4b80XEDGzCmEHuFz1iwCPcbMN1GS0UZlpJT4/QEiMqAi+tM4/YYTmfnwBwT8QfSQjjPFQec+HRl8/EAWfvYrQggOP2sINruV+R8uJCM3nSPPPZT0nFRKC0tJyUiJarf/oX154rvqqzvV/ISKw6FqZmF6JlKsdUBYuyBJB2r71OUC616xDTkhKLoV6TwCIZyI9PuN+PPQmjhthiC0HGHtjsRKzOrycTEWVnX/YkibAHq8m4JmRHzIQJW+nJB2k2FUS18iup6oSTuBZUhHKeglSJGDIBA2xBbIfgOZfxmE1hL7RiaRuh/pfR1KngVZClomMvVaNPfpMY5pXJq8Ic/blG8qoKSHdLb8ty2hNt6eODvaFSKNCIdWHXcmUVx0/yi2/LeNnz9ZjM1hI+ALsP/wfbjw3rMiDt176J588fq3NT6XYCCIQBhSq+Ex1BShCTr27sDGlZtrrGee037nDSiWERcCUrNSTZOqyrHYLeiBUJRcbIVbpYott9osdB/QlXP/dxoHnzyIOVO+JH9rIZom+OGDX1j/70YCviADj+7PUecegivVxTEXHs6LN7/BrcPvw2KzEAwEGXbREYybdGGt1kdqjKUrxmJWVd+rLSxslWQ0d+21t+2Dw6JRcTsA/6/gOMBItolrxMOUzUTmzgMmkLght2N8+AGMhcyQ4U7JOyX+YTkfIaxdjMpAxU9CYJFR7i3lMoQzHDeuJ3ID9UPZFCibgkQDdONfrRUy7SY01/GQ+Tgy73SQ5pMOrLsbui/FT1Jx49B3QNG96MKNVlkeoInQ5OPldt+/h2kEhMPtYN+j+yfURqzZs9VupaBS5IPdYeOOd8YzdekT3PbmtUxd+gR3vDM+KqPw+MuOwe6s+cKGDMlwXUlqHUMudUnrTjm07phT43DHISMHxXQRalaNo847lKlLJ3Hy1bF1J9zpLv4387q4mt9CGC4rhBF1c/QFh/HgpxMQQtBtzy6Me/IiDjn1AH78cCH+cChowBdg4dzfeOSiZwB4Z+JHfPjMXPxeP55iDwFvgLmvfMXrd82q0TnXFiEsiIz7MaRVy6+zw4hOSB2TlD6klIbmiQyBsw7p24HFoJtr5FfqrSKGm5LJCQ6wFMpmEN8FUxU/0YJYQeJHp1gRAoSwIay7oWU9gdb6O7ScWTuNOIC9plmz+s5/9a1QeBvS+xWy6K7w4mjVL7Ewil+k32/MxKNm/14oeaKGY2gYmrwhb9Uhh+PGHGUYhjA2h5WsNhkMuzAxKc79j93bNA486A/SfUDXqO1tu7ZmwOF9+WfBcl67cybfzPyRgH/nl3OPQT0ZevYhONyNU04rq00mj319F/2G7J6wMdcsGmv/2RDzBqIHdb59ez4FWwuwO20xZWT3Pao/mmYUcIiF1MFb6sPhsnPNc2O49vmxuFIj427fevj9qBjwgC/AT7MXUlJQyqzHPjSJEffz3pNzCAaCfP3WD9w/6gkmX/FiVHhoshDOoYicWeA6xZj1pl6JaPVRdCxzLdDL3kFuOxC5dX/k1v0wii7UMupBeoHYT1CAESJnC68dBBPNYvRByUO1G1NNEE4jlT0GUupI/89gH0LkjbWmeJElTxg3PvOeEK3mImzdY7vPQtXdMBsH0RgV4QcOHCgXLkywLBfGzGXe9O95b/IcSgvLOPjkQZw2/oQKHerqKMorZkz/8RRtL64ovux0O7jg3jM55Zrox6S8zflcOfhWineU4Cnx4kp1kpaTyuT595Pd1ghLCgaCjOpyGfmbCxI+j2TRe1BPXG4Hadkp/P7d3xRurf6R02ItXzytPmLBarcSCoai4uI1i+D1Vc9QtL2YKwffQtBffQxwZusM3tr4AgDb1+8gJcNNSkYKozqPZdv6HVH7O90OXvj9MS7c42rTUFGAfkN2Z8WS1XhLfWiawOawMe7Ji2JKKNQFKX1G7LGWEyUjW+s2vZ8gC24mYsYnXOA4NpzgsnN9JDnYwdrdkHO1HwCFt1G9r7kBES5E7nxDwpfwNff9ANKLtHaBgit3up1kACydwgbVbFZdXV9pVESoRL3nRmvzq/GktO1g8zBJa2+0VrNr1mcSEUIsklJGlU1qFoY8GRTtMET/F8xZTHabTE657ngGxnDN3H3ao/z4wcKICBeL1cKBJ+3H/2ZeD8CPH/7CA+dMiog735XQrBoyJLE5rUaShISrnrmEYRcegZSSC3pfxcYV1c9OrHYr1zx/KS/e9AaeYg+6rjNoxL5Y7Ra+nfVT1CJ2WnYqsza/yNVDJvDvz9Gx+TkdsiktKI2SKra77MzcNCWple710mlQEhZSkgGw74fIfKJC6rXW7W4bZr4wKdIg82XIH0WNMyZjUtlnHcMn3Kg4Ie1mIx685Bmj1J2+FbCD0CoZ3MrjdkH6bUYYYmi5yQJpHGz7gq2fEUYa4TJygPsMtPQJAOhlH0DR7URqqjgRWU9HlttrYFq8Ia8Jw51nmSa1WO1WPvHOAGDWY7OZcuNrCdWHbK443HZOvvo42nRpxeATBlY8jQDs2JTHhbtfbRquWZnytYTKsfs2h5Xd9+/Jqj/W4C3xVdwwHW4H1zw/hiPPPoSl8//lxqPuwe/xI6VEhP3t3fbsbJp85U53cev0axh07D7JOHWk7wdk/uVEzlxtYN8fLfvl6P1lOHwygSIF+pb+5hKtWMHaD4K/xjnaRkXVeWENGzEzo28BSw/D0NUpPr0esR2MSBsLWhZyx2nVhAVWPXZvtJy30IsehrLXSOzG5wyXmtsTWXCFUX4NK8Zn7AAtA5xHIVIvR2jZSO9nyOInILQBrN0QaTcYhZQbkViGvMlHrTQ0//21LirWvJzKfuGc9lm7tBEH0CwWuvbrxBFnRX95//z+XyO+PVx6DaIrIDncdlp1zGHj8k0RxwZ8Qf79ZQWPfHkHn7/2Db99s5S23XI586aR7HVIHwD6DO7NpB/u5Y173mbVb//RpW8nzp5wKu88/hH//rwi+tpLItZR6oosnUK0+yEA/l+Qoc0VhX1laBOycAL4fzRe24cgMu6JX/jX0sNIla+KSK/GiFvBvp+RHi4DhrERbii81sRVYA8n8zRRI44dMu5Det+D0heMhdWaoJcYN0/PW8Q24jYME6cbRTfSb0KEizmLrCnI4Gpk0T3g/wUoA70MymYgvXOh1RyE85idiVRNHGXIK7Ho89+4Y+QjMfUWKmcXZrfNjFm6zeawYbVb8JX5cbjtBLwB02QYiJ8NKTSBO82F3+una7/OLF8UK044cWpSbk4PhejWL1p1cMWvq3nkwqciQjo1i0ZGbjp+j7/iaea4S49mybw/TG94tvDi89XPxo4A6d6/K3e8PT5i24gxRzL/w1+iXCsOt72iCHNSCMUoRCJsoOeBpa0hKbvjNNC3U2Ew/d+Fy5F9EU4oCYFvLrLsHSN8znEQpF5l+H2rVpN3ngCeeIJWulFxJ/AbkIUUtkql0LRKx1kN90FoI8hqFkEbBZtxHQouhmC8eO5YOIybmb49dpaqyEJrE1UWoco+jrARr9x/APQCpGcWIuXCGo6r8VCGPIyUkomXPGdeHCLMPwtW8N9f6+jatxMde7fHapJtqVk0WnfOIRQIkdI9hWEXHUFJfgkzHngvShogNTOFEWOPYubDH0QZV4vNwtm3nUK/IbvTbc/O/PLpr0wa+0LMCkiJIAR027Mza//eELNUWTlWu5U+g3vTbc8uFdu2rtvOM9e8zI8f/BI1Xj2k4yn28MiXd9CqYw7p2anYnXaeufZl1i5dH3UjC/gCdE4wK7cy/Q/ty5k3j2Tafe9gs1lBGGO9f85tCcWXSxkE70dIzwcgnAj3GWA/NDprz3EglK0h2vcqjYVDMNT1ZCmRhre8HNkXSOcwZN4YCMzf2U7wV8AGKTdC2Qsgw7kQwgWaE8OnHctdpYfbDwKlcdzdQaNMnJYEed16IWBolEgPNV/UdYXjyy80bqqxLoKlnaG7bumC0KITyYxhhMu9RZWd8xrKj83IkDf58MOGIn9LAflbYlU4MQj4g7w90VixbtU+myEnH4DDFRmCqId0NqzYzOb/trH6j7W8ePM0+gzuTbvd2lQ8+ltsFuxOO+fffQbn33k6N7w8LiLkz+G2071/V8648UT2PmJPMnMzGDJyf6z2ut53BVmtM3jih3uqTTfOaJXOCZcfQyhkGGBPiYcrB93C/A8XxpzRaxYLhduLadU+G7vTuC6njz8BZ6ozIkzS6XZw6nXHk5oZ4wdWDWffdgrTVj/DtS9cyoQ3r+WtDS/QY+/q5UOlDCHzRyML7wD/D+D7Epl/NbL4/uid3eeFtaUr3xw0wIUsuhsZXGeUMDOLfpAeCK1Ber+GwAKibwYBKL1vpxEHQ6mv9CVjgS8peEFPpNhDIyGLqJkRt4LjaET6bUaxZ5GCLHowRhsaBJch885Bbh2MXvKU6VO2FOnmnx+aERnTjFCLnWHKij2cmntRtfKrfQb3YtIP9wFGCOKLt0zj3Sc+juuu6LFPNyZ9fy/zZvzA/A9/ZsWS/8jfXGBI5grBZY9fQEqGm4mXPEdJfikIGHziQAYN34ePnptL/pZCeu/XnY692vP2xI/My8MJ6NKnIxtXbEZKI8VfWAR6MHLf8nC9UEiPOysXmsCZ4iC3Yw6Pf3cP3739E89d/2qUS6MyNoeNaWueJat1ZFTH5v+28sr/3mLJl3+QkZvG6eNPZOjZBze4doX0foUsuIZo37cN0h8w4ouFw8jsK74Ho0KOn+hkFs0QZkq9FoofIzr70wkpF0HpyyZ9JUK5MW+q/u1GQKSitdkZ/62XzjSKX5he36qSAi5Exl2IKhV79OJnoHQSprP6nNlo9VyerTaoxc5qcKe5GDRiXxZ8vCimMbfZrfQ9aKcf1mqz0qFHO+xOW1w1xJVL/sPmsDHswsP5/LWvyduUTzAQqujHKNpcaWYhYf77C5n//s6b3Y8fGP+PlajjcNkZP/Vy7E47m1dvpWu/Tqz+Yy0PnT85IrJE1yU+j5+M3HTKCstinqvUJZ5iLxtXbOa5617F4bbHNeLOFAcjrzo2yoiDkWB182tXxjy2oZC+rzH/4Qeg6Bbj3whfcyx0w/fs/wnzn5A3rDFe29DUlmzArZQXdN6JDZzDI3cre4XYN0kTobOSF6IMOb4vMHfNOBGmlYSaLsq1UonxL13GHgf0wuGyY7FGXhohBI4UBydfE5m+/u2s+dVK2rrSnAgh2PzfVv5ZsDzaX5xAkk45sWb+TreDnvvuxm57deHAE/ejffe2HHD8vnhj1CEtzi+h1349qlWSDwZCfPv2T3Tr18U0KkQIQbd+nbnxlSuiNGmaHDLe01b5Z1ADI+r7hdj+7F0zv6B+cULma6C1CRejCD/5WDoh0m6I3LUmNTbB0EoBpO8n9B1no28dYoQVxkJLLNmwqaAMeSVSMlJ47Ou7uO7FsVGP/UITnHjFcFq1j0zPTsmMX3DX4bJzwuVGCFPe5oIk+LkjsTttOFMc3D7z+qjFPk3TcKY6TY9Lz07jwnvOxOGsXmZA6jpDzzkYh8uBVumJwGq30qVvR5779REOPuWAenOVSCljRhItnf8v9545kWsOnsD0B96lpCBOGJs1yWW4NKNau6KcuhZW8ELBhUY0imgFtoMg9fqwLEJmxV4ysJSauawE2Aaie+Yi88dA4Bcj2keaaTBpYO2MKF/QbiYoQ27Cx89/ETVr1kM670+eExVjvvcRsbWvNavGIacN5vy7zgCgW79OMdPOa4otLK87+sFzeH3V0/Q/rG/UPkIYN5+qmjAOt4NTrzuOZQtXxoyZL8di1Rh8wn6kpLuZvOB+9jlyr7CGuIXsdpkATB73IptWxwjXqwObVm/hlmH3Msx+Jse6RvHAOZMoyiuueP/Tl+dx41F38+2s+fz1w79Mu+cdLh0wnuJ885A7Yd8XIyokSTiON5e5bbEkQ1bABwRBXwOB76D4YWT+WGSl6yzLZlCzDFUJ/s+hcDyxn6Cs4dl/Z0TW87UffiOhDLkJq03Kh4HhAinaURyxTeoSzWp+GYdfdAQ3vnIFVpsxC3elujjvrjNwVCpUUV7suaakpLu58qnRjLzqWDJzY6eMX3DXGQw9+xBsDhvudBd2p41jLxnKaeNPoHWXXEMjPQauVCc57bO5/IkLAGjXrQ0PfDqB+z6+BavdwrZ1O/jvz3V8MnUeYwfcwJqlCdaCTICyYg9XHXAri7/4HT28MPvNrPlcsud1fDL1S3ZszOOZq1/GV+aviFP3e/3kbynk3UkfmzdqGwDWHsQuJlwDRAqk3Rj23TZO5fSWgQ/8C5AllSr2BP6uZVvx/N4apE1AtPoMYVIbtKmjFjtNaNettaket2a1kJoVGTKX0z4Lp8tOWZVUdbvLbhonffr4E+jUuz2zHvuQ/M0F7DdsAPsfuw93n/YYvjJ/RURKRm46fQ7szS+fLDaeDsLGymK1YLVbuOHlcQnFTVusFq59/lIufmAUW9dsp91urSsKNAw+YSCuFCfeUl9URuaIS4+m9767MeSUAyKUI6WUTB73YsS6QCgYwlPiYcpNb3Dv7FuqHVN1rPztP957cg5lxR70SuMKBULkbSrgqSunIiWmCowBX4D5Hy7k/DuNpyCpF4H3Y2RoC8I+ALJehvyLwqXVaosLMp9D0yzIjPuQgb8hVF3dTUXt8UHZdPTgKgiuAhkvTNiGsVha0wXjAIRWN+kqQPFQhtyE8+8+k7tPezTCWDncDk65dkSUNvkBx++L3eXAU+KL8ONarBpDzzYX1xl8/EAGHx8ZQTR9zXN8NeN7tq7fQb8DezNw2AAsFgulRWVs35DHr1/9wZIv/6RNl1yOv+wYOvZsV6NzSs9OIz07LWKb3WHjie/v5YFzJrF88WoE0HmPjtz8xlV07WseR1tW7GHLmuhq8lLCH9/GnikFA0HeuOdtPnpuLt5SH/0P78tlEy+gY6+dSSveMh8TjnuAf35eQTAQjOmGqppYVZXM3HRjTIG/kHnnhsMIPcgyN1h7QtotkD8a8xmawHC/xFus9EL+GGT2awh7f6Spr3UXwroPBGNJvzYQshB8n1WzkwVy3oYdp1BzQy7B+yUy9QqEiL/u1RRRceQx+HrmDzw//nXyNuXjTHFw2vXHM+q2U0yrs69ftpF7Tp/I+mUbEUKQ3S6LW6dfze7792yEkdeM7RvzeG/SHH7/dikde7XjrFtOpvPusR8tg4EgJ2acZ2pM23TJ5Y3Vz5ged88ZE1nw0aKKzFQhBO4MF1P/eoKcdoYY1+QrXuSTqfMqpIarQ2hGqaXK3+Hyhd/9hg1Abj/apBKODeyHGotdwT+pU6if1hUyH4e8UTQpWdhk4jgRgkvD4ltNjcqhok5IGY2WdhX6jjPCRZ7NbFs8BUgbuI5Hy6htbdH6R6kf1gIpJX6vH5vDZmrAq7Jt/Q6CgSBtu7ZuFo9oG1duZtx+N+Mr8xHwB9EsGnanjXs/uoX+h0Yvnpbz+KXP88Xr30QYc6fbwUX3n8XIq6KrC21avYXRfa+NMv42h41Trz+Oi+4dBcDxaefiLU08ftfpdpCWk0rxjpJwceoA595xGmfeNBIZXIfcPoL4s+4AdY/ZTiTuvLliwSjkkPxi48nBBra9QUtDuM+uUCaUwbWG3o3Mq7J/Ip+VHdFmEUI0zXUPlRBUC4QQOFwOyoo9vD95Dt/MnI873cVJVwznkNMGRxnr3Er1P+sbn8dH/pZCsttmVqTD15QpN75BaVFZhX9cD+l4S308cenzvPT3pJg3o8ufuICiHcX8PMeober3Bhh+yVBOvGK46f5r/lpfsV9lAr5AhCRtojPxcnQ9xNO/PETB1kIKthbSc5/ddqb9C43YMy9J8uK8d1UjDoavuakacUDLRKSciyx5Flk4HmkbgEi9DmHrBa2/RpbNNErVhf7DiEm3xpAProIsDcszNB+UIa8Gn8fHFYNuYct/WysM0Yolq1k6/18ue7zhRXV0XeeV29/k3UlzKhb7Tht/Auf+77QaPwUsmfeHaYLRptVbKSsqM61aD+BwObjj7fFs35jH1rXb6dirXZT/vTIderY113e3Wdltr84Vrwcc3pfFX/wR4SoRQrDXYX3ouFsZX0xfid8r0CwSixXG3rWJzOxCslp3jmpbWDogLZ0gtJKmV0xBkRQseyALb9xpnH1fIf0/QfYshK0nIuU8SDkPAKkXIrcOrr5NLQdEVvX7NTFU+GE1fPnGd2xbtz1iNukt9fHR85+bliqrb2Y+8iHvTZqDr8yHt9T4m/XIh7z35Jwat5WSYV5NR9MEtgRm+a3aZ9PngF5xjThAp94d6Dtk96hQR5vDyklXHlvx+oqnRpOS6cYeFiKzO22kZLi5+pkLuOreL3h41kpOvWwbZ121lWc//5cR526ODEurgsh6EkQGhnsgFkkIRVQ0AgKCi6rMsCVID7JkksnuLqr/rB2I9DsrJkRSSqR/iVFgIl4WaBNAGfJq+PnTJaYaI1ablaU//tvg45n5yAd4q0jtest8vPXQ+zVu66Qro5OFbA4bh552oGmx6rpw57s3cGQ4nl1ogp777sYj8+6kbdfWFft07NmOV/59krMnnMLBpxzA2RNO4eV/J9Gxuw4yxO77lDF6wibOHb+Fjt39GPrcv8TsU1h7IFp/A+l3YW7MXZAyLmzsFc0LGaMYhQzrtUcihB1cIzGP+U8F+5GInGkIp1HQXYa2Ircfi8y/EFl4M3LbMeiFE5CyabrSlGulGnI75kRUwSlHIslqm9mgY5FSmsa3AxRur74Ac1VOvmYE6/7ZwJfTvsPmsBHwB+k3ZHeuemZ0XYcahSvFyXVTxnLN82MIBUNRYZzlZLRK56ybh4PvO5B+pHUTsmACMaNCqimILIQL4R6JtOQg86+gorCxMEIRReolSMch4cgTsyxNgTGTS7AmpKLxsZjrsIv0CUhZYujIC7tRlMJ9ASLtuii3pCy4OuxbrxQC65mNtO2FcJ9ef2OvJcqQV8Pxlx3DJ1O/JBTc+SMXmiA9Jy25FWkSQAhBx97tWf9vtGBQlxhx3/GwWCxcN+UyzrvrDNYuXU+brrl06FGz+PSaomkamj32g6D0zUcWXBYWuAqF/2L5/l2I1EsT6lc4DoHcT8OVerYaEQ6OoQhhRdj3QrftaRRjiDDmLsj5ALzvG4tmshQj0kX53BsM6yDIuBN2nEBiEgBOROo403eEsCMyJyL1PAhtNtLxTcSxZGhb+LtQNY/BA2VvQBM05Mq1Ug2dd+/ArdOuIS0rBVeaC4fbTte+nXjkyzsSCklMNpc/caGJdoqdyyZeUOs2W7XPZp8j96p3I14dUi9F5l8aFvuvrANuZjhtkHYTwnFYwu0LS3u0tCvRMu4x6jFWKpQssl4E18k7fan2gxCtPkCzdUVLuwatzQK0tn+C63Ti+9wVScXaEYomkJARF2mQfodx0463m5aNsPUxNeJA+PsXw5+uN8XSeWpGnhAHnrgfs7ZMZfUfa3GmOmucVZlM9jtmAA9+OoFX75jJ2n820LVvR86/6wz6DG56Ivg1xdALTzAs0HEIWsqouvUXXBUuvrvACDdznYJovQAhYhtqkX67cVvxvI9xg1GiWTXG0teoXhQ31T6M92ugajx4DHLmoVmTsN5h6QRaGugmBUicR9e9/XogKYZcCDEMmIRxG3tRStl0U6NqicVqSaicWEPQb8gePPLlHY09jOQTWk9ibgs72IfUqSsZ2hFOGik2+pRBKHsLGVyJyH455nFC2CH9bqR1dyidDnpTzHhs2oi0q5CeWeHCDtWRT0LfCduQ5BhxQAgNMh5C5l+GsTYSBJygZSfsymto6mzIhRAW4GngKGA98IsQ4kMp5dK6tq1oYdgPSHDHANH+y5ohPTPCRXcrGwkf+BchA8sRtp5IvdTYJrIiFsPkjpMgWFsFPoUsuAxEDkZ2bbwnmnhJXZVwHIvIfKza3XTPHCh53ChKYemMSLsB4TzCdF/hOAhafYgsmw6hdWAfjHCdHNsd08gkY0a+P7BCSrkKQAjxJnAioAy5okZo9v7oIgdkdfH5EkomI91nY8wjaoH/d2K5cWTBlUh9S0R4m8QJ9v1AtFZGvM7o4cLTgvjaJxpxo4VSbjTcM/paZNmb4DoJoZknsellH0DR7VRINoRWGvVbMx9HOIeaHiOsXRHptyZ0Ro1NMlbrOgCVhajXh7dFIIQYI4RYKIRYuG3btqpvKxQAiJzpINIx5hhxvp7SW1G+q6ZI6TeEoEzxQmiVSYyyF/zfge+dWvXZMhHEnytKIJ3oz1mA1hss1axFlU6EspfA+4lRgGL7CCMixYySx4jW3fEiix+J30czIRmG3Cw2LOoWK6V8QUo5UEo5MDc3NwndKnZFhLUbovWPiMzHEWm3gmW3WHuCZvhEZdCYkUnPR0i9akX7SKT0IAtuAV1NJuofGyL7NXCfT+wQUj/YBkVvtvcC19nEvxEE2RnN4gF9m2mmr5Qh0DebNxEyLyLT3EiGIV8PVA5i7gjUsDKqQrETIexGeGDKeeGiu1WjSJzgPgOwoxc9YMzEiu5HFt2O3DYE6TfXztZ1H3L7SeD7CBUL3hCkGe4L/2JihmxaOkCg6uclwfsF2PqCbd8a9BcA7+dRW4WwGBoqpv03bshtskiGIf8F6CmE6CaEsANnAh8moV2FwvBfpt8OIhMjvdoJ7jMRaTeB/3vwvInh6/Ya7hBZEq7xGBl3LIPrYPsREFpN4xjxpi9rnHzywrrvf2DMnqvOrp1gPwhzM+QF72eItOupUZ3VWKGjKVcCVQtGOCHlmsTbbsLUebFTShkUQlwBfIYRfviSlLIudbQUigg092lI18mg54GWYYQAAnrZrBiypAHwLwKHEQUjpUTmX9SI7hQrWDqGU753cbSugAR9LZE3zABGGbYMoNC4Hqk3gb4dc/kFYcT22/Y06qNGFbkuj2ip3IcL3EZugZQeZOkM8M4BzQ2uUUaN1dKnwt+jXEi9Ds19fHLOu5FJShy5lHIOUHP5PYUiQYSwgGXn2oqUQQjEmi8IIsLagn81vk88lLzC1E0XDZE2Fln8BOZPPQHAA7a9EVnPIrRs9IIbY7QlEa7jjc896ylk/iVGPUF8gMOIIAquBbkVEEYegPMIhPtcpPQjd5xl1PcsX+AM/Aau09Faz0fKAEIkVxSusVGZnYpmiSy6M/YCFiHjh16+b6jh5YYjaSmCW5oxe7a0i/PZ+CHwBzJ/HCJnhhEJZIoAS1vjf/b9IPdrIzpFLwD7IKMyEEBgoaGbYuuHsBoJe9LzEQRXExGlIj1Q9iYy5QKEJXYpw+aKMuSKZofUC8Mp8mb6GxqkP1BRQFd6v4CCW4D40SyKZBBEFj8KKaMh8DfmZfaM/Qj8iQyuJ7ZujYXKvnGhZYL7rOjdKt2wy5G+bzF11wgr+BeCa9cz5Eo0S9H8CG0yfpRmaO0QzqOR3s/Ri+435EhJQNNDkRxCa6D4UUi7PpwPEGORV9iMmpruM4k25lawH1D7LEqtNTHnqFp27dps4qgZuaJZIUMbkQU3hBXqqqKBdXfk9qNBzw/vo8IMG5yw9IFo/ZMhSuZ5m+inJx2svRDWPZCB3w39eRGeV1raIzIernX3wn06smwakS4tYWjQ2xMo99YMUYZc0WyQUiLzzo+zcOgwjHdoM8nzSzswFk7VDSFxvBD8x5AJTrsa6fsC9EJ2LkC7IPWmCpVJkfU0MrjCcMdY2oNtnxrXn62MsHaDzInIwpswimProOUisp6PkC7eldg1z0rRbJFSGroZZS+HF7YOQKRdj7B2MSIP9G2YVq4XrSDrWcg7i+QuLtZNnKtl4gJrX8DQ/qbVbGTpK+D7BrQ2iJSLEI5IgTRh7QHWHkkbgXAeCY75xs1BuIxqUHW4OTR1lCFX1Bkp/eD71pCEtQ9GhKMNatVW8YNGkk9FZfS5SP8P0OqjsBGPsawj86HgKpJueIU9hhtHEZsQOE+qeCW0bETadZB2HVJ6iOU3l8EVyOLHIbAEtNaI1MsQzmNqPQoh7GDvX+vjmxPKkCvqhAz8jsy7GGMWbOh6y5RL0NKurnlbegGUTSdSlVA3KqOXTkWkjDFJDCknBPomEpY+TXhQMWqFKuKgIQI/g2WnqqAM/ocsvDlcGFkg7YMQGQ9U3PRlcCVyx6nh6y1B344svBEZ2oyWcn7jnEYzQkWtKGqNlEFk3iWGlKgs3VmirfQlpG9+zRsMrjBmwNFvgH8RwtIa3OeEy7HFoo5Vzm0HE1nmS/nGa44X6dmpEin1UuSOM4yZNiGMz3M+cscZFVIKsniyoWhZ+XpLD5Q8YTzxKeKiDLmi9vgXYl4YwIP0zKx5e5YOMWbcGoSTPUTaTYiMh8JiSrFikJ0YqeC1ILAA5RdPArLSOoX342gjjQ6yCHxfGy8DSzC/CUsIbai3Ye4qKEOuqANeYsYJVyMna4awtAtXCao6K7cjUkYjpW784IUbkfUCuI7DvEiuFyMhpDaLW2r2V3ccCNdJFa9k8D9ME3Skb2cEUqxsSxkErVWyB7jLoQy5ovbYBkbOvCpwI1wjatWkyJwEzmEYxtxuJPhkTQZhR247HJl/MbLgGuTWA0HkYggxmaEMcu2xAXVInBH28GcYfmnrC7hNdgwiPbORwbWI1MsxlSt2jkBoabUfSwtBGXJFrRFaKqTfjfEDDM+MhduIFHAeW8s23WiZjyLa/IJo/S0i92uwH4zMu8jQ7whL1YLPCFFMuwkssYpiK/927bBA1tPU2jzIYmSokkiZ86iw4JlJbEVwKXL7icjQRki9OixX7AQc4DoOkXFP7cbQwlBRK4o6oblPQtr7IcveBlmIcAwFx+G1r6VZgRNZ/FhYbzzW7NoHgd8QGQ8j806rY3+7Mg5i1Sc1JwCWNmDtB8Hfa9fl9iPQLZ0RmQ8jbHtCzkyjMpN/XpUdJVAKRXcbL12nQMpYhJaB0Mxm8Qoz1IxcUWeEtQda+s1oGQ8gnEcmwYhjVJbxvEZ8F4kEPc+IPW82NEZSSk2MOEAI8i4IV2GKFyEUj4BR4DjvPGRoO0LLChc5jtWe3/jzvI8ILFJGvIYoQ66oFimDSM8H6HkXo+dfjvR+hZT157bQdQ/4PklsZ/vgOLrkTZFm8hCsbwWtA9j3MdxltUUGdoYiJpS56UGWvVH7/loozeRbpWgspNSR+ZeGQw2NyAPp+wHcpyHSJ9RPp8F/Et+35EkQWSA31c9Ykk4AY1beFPz38cYRQsityMwXoHQyeD4CfX0t+vDvrIxkG2AoE+pr4h8iS2vRT8tGzcgV8fF/B4FFRIaPeaDsrXBYWT1g6VqDnT1g7Z7gvlZwjKB28xcNcILWFmxHUKM6klE0BSMO8a+DRFr7QeFVUPZaLY14GK0zgKF1knppNf06wDm89n21UJQhV8RFer+NoTUiwF+L7M0E0CxZYO2Z+AGBBTHeqOqPtiLSrgX35bUYlQVc54LrdAj8WKXt5vYzEoaolciMvYt1IELfYnzGddKaEWDrtfOV8wjMY/8B7GDphHCfV4f+WibN7RuoaGi0TExjtYUlXDhgJ1LqSN936MWPIUtfQ+p5te83exZYEjXmMWa4IhVwGuXHRCYiazLC2hlEnDR+kRs+rioB8H8FZVMwEo7KFxDtNRhnU0FCcBnIOHVMLVnIkkkJGHFB/AVcG8K21869tWxIG48RYlhufqyGyyX9DkSr92pfUKIFo3zkirgI10hk6RSiCwMIcB5R8UpKPzLvQqPQsSxD4oSSiZA1FWHft8b9apobcj9G6nnI4BajovqO4UQbbQsxZWut/REZE4z0cGuvCi1qYd8LWeokuhSZgJQLofRp8/ZC64iOAPFDaDWQDhQlfoKNjlmZvEr4Pqd6OWAn2PqDYziU3BljlxMQlYpmA2gp5yPt+yE974IsNRQO7YcghJpX1hZ15RRxEdaOkPGoEbkgUsOz2yxE1ssVdTEBZNkMCPxRaQbnNQx6wdVGan1t+9ey0ex7oNl2g8yXKs2WBYgccF8W+2DNgbDuhrD1iSwoYD84HEFR2c9tBWt/cF8YY0YOccP40h+J8YaV5M2XEg3rtBpZt3UinhHXwLIHpN2IyJ6K0OyYZ25awdbHtAVh64OWPsEIWXUcpox4HVEzckW1aK6jkc5DwL/YqLVo2zu60ornfUyL7coSCC4HW++6j8N5ENKxKDwzFghrJ2TZ20ismBoeYa7RIYQFcqYhS6YY4xYCXCMRKaMRwoLMmmJUIpIFJLYwqUPRzUQn3lgw5kpJkAtwngzBtRBcmMDOdgjtqHuf0YMA4UBkPWNUtg8j9R2Y3+SCUBf3miJhlCFXJIQQTnAcGGePWAUfpOFPT9o4BFg779zgGBKjbxfCFbsogRAuRNpVkHZV9Hu23ZEZE6FgLIkl0wSB/BjvJUPzRYD3M0yFp0wpA311LbtKN1QJTQlB2u0RRhwA+/4YTzdVxifcCPv+tRuHokao5xlFcnCdjmnWnpYDlkTDA2uOsLSF1MvDfYcX3YTbMPD2gyL2laEN6IX3oG8/Bb3gJmRgWVR7emAletF9Rux0ncMEayOHaxbWGE5jr6vWOmBco1g/e2e4OHEseeAAFN8ervJTCdsAcAwm4vMXLrDtDfZBdR2wIgHUjFyRFIT7FKT/a/D/aCgiChtgRWQ9ZVorUUoJvs+QZdONAgLOEQj3mQjhRMpQOH79b7B0BOfRCOGI2beWejnSfiDS8zZID8I5AhyHRfRrVKA5LVzhPQDBv5DeTyHrBYRjEFKGkAXXgu8zGjXO230x+L8NJ0Xp9TAWASnXQelEom8M0hAhCywG3cRNBoDFcLE5dt4khRCQ+RR43kN6ZhntOk9FuE/ZpetkNiVEfaZax2LgwIFy4cJEfH2K5oYM/A7+X0DLBceRMTUz9MK7wftOpVJqTmMBMuslyD8PQmuNaBPhNB7Rs980QgcBGdqK9LwHoU0Ix2BwDK22OrqeNwb83xBlGC3d0HI/Qy99HYrvJ/4s2lLN+0kg9Ua01NFI6UVu2R/TdYcIUoAyEjf4AlxngpYNpS9inE84w1O0Bi0VQiuJeZ4iBZE1BWGv62KqojYIIRZJKaMuvpqRK5KKsO0FleKGzZDBdeCZRaT/2QvBf6HgCgiuosK3LMtAepGFNyFyZiD9vyDzR4MMAX6k931DxjZnekQUTRSBXzA1dqG1SL0EyqYR20gLsPYB5/HgPAa2D6d6AxuL8hlqDMPrXwCMNp5MtIw4M+NyPLHbMkWC511Em9/BNQLp/QI870FoI8iN1d+nhAtp7QvSF/cpybRn6TfK+WnZdSrQrYhG+cgVcZGhzegFN6JvGYS+7Qj00pcN10ddCPwSYwE0AIGfiV4g1CHwG3qoyHB/SA8Rhj64Aln6Wvw+qyQvVXoDueM8CMXT/7BAxgOIlAvQrB0gdSy1VwWUGNEtMais451yaQLt1cZv7gckwtrDiPEPbSSxRV07aB1h6z7ILQPQ8841bsqJjLJsJnLrIGTeOchtR6HvOAepx1ogVtQUZcgVMZF6PnL7SeCdDTIfQuuh+HFk4a11a1jLplZyrqH/wkUlquIDz+z4x7rPJ3oRT8OoCfkn8aeiIdhxGnLbIejeT8B+qFEEwdKJ2j3Uxpll23ZmiQr32WA7IE47GrF/whZiVk+yDaiI25aej0nMiFuNm2HwD4xrFQL/L8i806MXP6uge7+Borsii4IEFiPzayOVoDBDGXJFTGTZjLASXWUj5wXvx0ZFl9piP8jwfSeMBra9EFoGxEouErFKvoXfTrkAXCMBO4g04190EvN5S8BnSLsWXA15o4ysVa0diSfpJIIAx+FGj77vkTtGQvA3oDXRCTcWcJ4Alr5E/4zLVQ1t4b/ySBi74eNOv3Pnrt6P4ozHFk4Ay4CUSzDcOJWvlw66B7yxJYdlcC0UXEZ0JmkQAn8mPKNXxKdOPnIhxCPA8RjPaiuBC6WUBUkYl6Ip4F+I6WxN2CHwD1ja16pZIWyQ/Tpyx7kgYyWuWIFQWAvbgch42JgBW9qHU+Ir+4UdxgJe3D41RMZd6O6zwPclBPPAOw1T14TIAms3CPyOeYZjeAYaWGh+fK1xIByHIX1fI/OvYufM3QPYwXGCEUNv7YKw7WMkRIU2IHeMMmK/ZSh8jAz/lRnHaW3A2gVseyDc51b4p2VoM8ji2MNJm4Cw9QFbXyh9GSnNZu5lyOCamM9XxtNbjCxRYQN9B9CpmuuiqI66LnZ+DtwipQwKIR4CbgFuqvuwFE0Cazfw/0TUD1GGam3EK7B0hJzZsP0won3iTkgZY4SuWTqC8xgjIQkg62nk9rOBAnYa0QD4Pke6T4q5ACelRBY/DGWvGzci6SWmEXaOQLiOReaPieHKKSeWEbeB1t5Edzue/rcALRtZ/Aj4vifa/eIH34fgTwGskP26cZSlA+TOA/8PyNLpRthmxOzXD/o2SHsOYe0WGd0T2mC0FUN3RbiOMUSuAGnrHb5uVY2yG2Hb3fR4Kf3hm10MZLBmKpeKmNTJtSKlnCtlxSf7E9Cx7kNSNBUMOdGqLgsb2HrF/PFWhwz8hb59JHLLANh+KFj3wFj8s2IYOhc4hyFSxxl/rhN3GnFAWLuD6ygiXRo6+BcYNT5j4ZsLnumAP2yc42iJ2PoaBkZWIywVEysiZxpkTCJyUVQjpt8aCfpG8LwVv/CCLAVZiCy4oqJKkxAWhOMQEBJzoxyCvFORW/qibzsS6f0qPMw4iVqWHhVGHAD7kPCaQBV9GosRZmpONcqIadcgtJTY7ysSJpk+8ouAmM4yIcQYIcRCIcTCbdviyGcqmgzC2gWR/SJYOmP8gG3gOBSR9WKt2pOhzci8cwyFRHSMxJy/DYOScjm4LzJEmDIeMiqxe2YjPe9FyOFKKcHzIdEGyweetyvt50d63kfPvxK98E5kyfOVYtbj4URYuxv+ePd51Dw6xYbImgzBlVB4NZFp6yHjvEVm2GXkItrHXp3iYHlTW8MupkpY98A8MzQQPndphFsWXI30LUBomeA+m+goGhtkPR+xRQgNkT0dXKcZi54izdCnyZlpuMpMEML4vkQ/+AuwH46WclFi56qolmoTgoQQXwBmQZ+3SSk/CO9zGzAQOFkmkGGkEoKaF1JKI2oFR51mUHrxRCh9iWhXiguRM8PwxwK6Zy4UjmfnPCME6f9Dc59muEi27IG5W8OK1nYpUvoMv3FwBYnrk4TR2iJyv0YIzejL844xZn1H2J+sx+g7jPtShOs45I5TTM4zjP1oRNqVSL0U8s+lWklZM4QLkfMOolIdTBnaitw+rBp3UBjbfmg504xzLJsOZS+BXgj2/RFp4xHW3Wo+JhNkaDsy78zw9QuAsIKlGyL7daU7XgtqnRAkpYz13FTe8PnAccDQRIy4ovkhhACRXf2O1RFchqlxExYjk9PWx5h9F44nykdcdDfSPghh7Yy07WPiexUQFnOSZe/WzIiL8M1JuBBZL1aE5gkhEO5TwX0qAHpwA+Sfb4w1FpZuyLxR5udZjr4ZYettJDrFcj2IbEDGVmAUGVEaNsLSGrLfRBbdGS7PZ6MiVLAq4dm8EAKRcjaknB17vHVAWFpBq88M331wDVh7g32QSt1PMnVyrQghhmEsbp4gZZ3qQSlaArb+mCbDyABYw+XAvHMx96uGkN6PAYzwOZHKTn+zHUQqIv32cBtzSMyIW8FxnOHjTbsNkfstolJZsqoIfZOxcBiP0skgq1E8dBxqtCec4fqUVa+JC9KuQ7SeD7k/gq1yJXuj4pHIfNLUGApbL7Sc6Yg2f0PrRUZBDjOssc8z2Rg+/MMQKecjHAcoI14P1DVq5SmMb+Hn4Q/nJynl2DqPSrFLItxnIEtfCS8ilrsnnOA4aOejvPRj7rrQw4JXhrGi1Rxk2TRDWMvWD+EeZcxIAbS0BEcUBN8XQAi8nyJD64yanjGQvm/C0S5x0DcSP2U+FZFybsUrkXE3UpYYkSrCZlwb99kI12nGbNmSg8w2olGk/xeE1gZcxyG0rLjDEEJDCAd6yqVQ8hyRNzYnIvWa+OehaFbUyZBLKXtUv5dCYSC0bMh5B1n8UDhMzgnuMxGplTL8HIdBsVm1HTvCOXRnW5a2iLTrzftxn430/UBis/JKhrn0FaTjCIS9f4wTSMN4Cog347ZjPOia9G3tDVkvG4uM5U0KFyLrWSOmO7QZqbXGmBPplC+ECqEZi8zhmXxNECljkSIDSp8z/NTWXoj0WxD2ATVuS9F0UaJZigZFWDsaUR0x3++MTB0DJVMo1wQBJ7hORtj2TKwPx0HI1NFQ8jQ1S9jxIT2zIbQeWfIU6FvAugci7QaEfQDCdTyyJPbYwQHuM41sSd1XqW872PZBy4mjByPSkWUPgfdzJJohTpX+PzTXiBqM36RZIRApoyBlVJ3aUTRtlIytokkiA38gPR+CDCJcx4JtYI19q3pwLeRfEBaiqpSVKDLCLpKqmYoCrAMh+CdRrojs1xH2/uiez6DwRmOBVpanrLsA3dBAz3wM9O3IogcNXXEcxmJpyliE7yvDPWNpg3CdAWjg+wqEzTjXwGIi3TJOIxyzakUeRYslVtSKMuSKXRop9XCRij+RIseoHCTcsO1Qog25I2ygTdbt7QeiZb9itKmXGXKzQkNqHRH6RiOkzmqeDyelB7njTENhUZZhPAhLdopeSWKHKh6Klj2lFmdev8jgSmTRPeD/2dDNcZ2KSLu+xtK2ipqh9MgVLZJy/zKOQyNiYfT0u6HodowEnHIXSMgkBT1M4O+dbWpucBriVkab8UvZydLpEFzNTn98eR8JCHYFV1S/TwMjQ9uQO04Px6tL49+yGcjgKiOBTNHgKPVDRYtEc48M631X/gkEiWlcLXVQn/DOodaFKKxNT1DKUMX0EekG8oH/Z2RwZWMNq0WjDLmi5eKZQWIp8U5E6hW17yde5aLqcF9c+2Pri8AfmCd2WZvkE0RLQBlyRctFjyWhC0Z6hA20VpB+DyLsSqkV7tOp+U9NMyJdnDUPOax3bH0w1XSRwfhCXIp6QxlyRcvFEktPRGDEcDsh7Q4094l16ye4CvNsVYHxE7QCdrD0NvoUmeC+EJH9atxmZXAdMrDUkIttQIR7lCFpG4ED7PtEaL8oGg612KlosYj0W5H544j2X5cXZQAKb0DautfNQHnextz3rhmuEy0F4RyOsHZNqDkZ2myUSQsuN9wZaMj0u+scc54owtKmiqaLw1BCTL+5QfpXRKMMuaLFIhwHQ9YUZMnjhqCXLCM6gSiALJuxU8elNsTUNReI1DEILVZhaJOmpETmXQShVVSWLaDwFqS1W4WCZH0jbL0QOdORUirtlCaAcq0oWjTCMQgt501Exv0xFiVDENpUt06cx2A6Z7L2rJERByC4NFz1vuoNx48si5M5Wk8oI940UIZcoQCwDYgxc3aBvW4LjiLtGrC0raJgmGYU0Kgp+nYQZj9bve43HEWzRblWFAoMES7pPhs8b1aqJOQASzuE+4S6ta1lQ6s54JmDDPxqZIG6T6pWwdAU254xbjhOsB9Sp3Eqmi/KkCsUYUTazWAfgCx93chWdA5DuM9D1CUOvLxt4QT3yQhOrls7WjYy5WIofZmdejB2sOQg3GfUeZyK5oky5ApFGCEEOIcjnMMbeyhxEalXG9WUSl8FvQCcRyNSLlCl01owypArFM0M44ZzNMJ5dGMPRdFEUIudCoVC0cxRM3JF0pF6PrLsPQitRtgGgGuE4SNWKBT1gjLkiqQiA/8aVeRlAPAaFXdKJkOrd43oDYVCkXSUa0WRVGThzSCL2Zn2Xgb6VmTxxMYclkKxS6MMuSJpSL0Egv+avBME79wGH49C0VJQhlyRROJ8nYSt4YahULQwlCFXJA2hucF+AIYEbGUc4Dq1MYakULQIlCFXJBWR8aBRFk2kAE7ABba9EamXN/bQFIpdFhW1okgqwtIaWn0G/vkQWgfWPcC2l1LJUyjqEWXIFUnHqFx/UGMPQ6FoMSjXikKhUDRzlCFXKBSKZo4y5AqFQtHMUYZcoVAomjnKkCsUCkUzR0gpG75TIbYBaxq84+TRCtje2INoAqjrsBN1LQzUdTCor+vQRUqZW3Vjoxjy5o4QYqGUcmBjj6OxUddhJ+paGKjrYNDQ10G5VhQKhaKZowy5QqFQNHOUIa8dLzT2AJoI6jrsRF0LA3UdDBr0OigfuUKhUDRz1IxcoVAomjnKkCsUCkUzRxnyWiCEeEQI8Y8Q4nchxHtCiMzGHlNDI4QYJoT4VwixQghxc2OPpzEQQnQSQnwlhPhbCPGXEOLqxh5TYyKEsAghlgghPmrssTQmQohMIcTbYRvxtxBicH33qQx57fgc6Cel3AtYBtzSyONpUIQQFuBpYDjQBzhLCNGncUfVKASB66WUewAHAONa6HUo52rg78YeRBNgEvCplHJ3oD8NcE2UIa8FUsq5Uspg+OVPQMfGHE8jsD+wQkq5SkrpB94ETmzkMTU4UspNUsrF4f8XY/xgOzTuqBoHIURHYATwYmOPpTERQqQDhwBTAaSUfillQX33qwx53bkI+KSxB9HAdADWVXq9nhZqwMoRQnQF9gYWNPJQGosngBsBvZHH0djsBmwDXg67mV4UQqTUd6fKkMdACPGFEOJPk78TK+1zG8bj9bTGG2mjYFa3rcXGsQohUoF3gGuklEWNPZ6GRghxHLBVSrmoscfSBLAC+wDPSin3BkqBel9DUqXeYiClPDLe+0KI84HjgKGy5QXjrwc6VXrdEdjYSGNpVIQQNgwjPk1K+W5jj6eROAg4QQhxLEbF7XQhxBtSynMaeVyNwXpgvZSy/MnsbRrAkKsZeS0QQgwDbgJOkFKWNfZ4GoFfgJ5CiG5CCDtwJvBhI4+pwRFGRempwN9SyomNPZ7GQkp5i5Syo5SyK8Z3YV4LNeJIKTcD64QQvcObhgJL67tfNSOvHU8BDuDzcHX4n6SUYxt3SA2HlDIohLgC+AywAC9JKf9q5GE1BgcB5wJ/CCF+DW+7VUo5p/GGpGgCXAlMC09yVgEX1neHKkVfoVAomjnKtaJQKBTNHGXIFQqFopmjDLlCoVA0c5QhVygUimaOMuQKhULRzFGGXKFQKJo5ypArFApFM+f/y1o6j9lZLFgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "predictions = model(inputs)\n",
    "plt.scatter(inputs[:, 0], inputs[:, 1], c=predictions[:, 0] > 0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x22543e3c910>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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 = np.linspace(-1, 4, 100)\n",
    "y = - W[0] /  W[1] * x + (0.5 - b) / W[1]\n",
    "plt.plot(x, y, \"-r\")\n",
    "plt.scatter(inputs[:, 0], inputs[:, 1], c=predictions[:, 0] > 0.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "## Anatomy of a neural network: Understanding core Keras APIs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "### Layers: The building blocks of deep learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "#### The base Layer class in Keras"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**A `Dense` layer implemented as a `Layer` subclass**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "from tensorflow import keras\n",
    "\n",
    "class SimpleDense(keras.layers.Layer):\n",
    "\n",
    "    def __init__(self, units, activation=None):\n",
    "        super().__init__()\n",
    "        self.units = units\n",
    "        self.activation = activation\n",
    "\n",
    "    def build(self, input_shape):\n",
    "        input_dim = input_shape[-1]\n",
    "        self.W = self.add_weight(shape=(input_dim, self.units),\n",
    "                                 initializer=\"random_normal\")\n",
    "        self.b = self.add_weight(shape=(self.units,),\n",
    "                                 initializer=\"zeros\")\n",
    "\n",
    "    def call(self, inputs):\n",
    "        y = tf.matmul(inputs, self.W) + self.b\n",
    "        if self.activation is not None:\n",
    "            y = self.activation(y)\n",
    "        return y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2, 32)\n"
     ]
    }
   ],
   "source": [
    "my_dense = SimpleDense(units=32, activation=tf.nn.relu)\n",
    "input_tensor = tf.ones(shape=(2, 784))\n",
    "output_tensor = my_dense(input_tensor)\n",
    "print(output_tensor.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "#### Automatic shape inference: Building layers on the fly"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "from tensorflow.keras import layers\n",
    "layer = layers.Dense(32, activation=\"relu\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "from tensorflow.keras import models\n",
    "from tensorflow.keras import layers\n",
    "model = models.Sequential([\n",
    "    layers.Dense(32, activation=\"relu\"),\n",
    "    layers.Dense(32)\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "model = keras.Sequential([\n",
    "    SimpleDense(32, activation=\"relu\"),\n",
    "    SimpleDense(64, activation=\"relu\"),\n",
    "    SimpleDense(32, activation=\"relu\"),\n",
    "    SimpleDense(10, activation=\"softmax\")\n",
    "])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "### From layers to models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "### The \"compile\" step: Configuring the learning process"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "model = keras.Sequential([keras.layers.Dense(1)])\n",
    "model.compile(optimizer=\"rmsprop\",\n",
    "              loss=\"mean_squared_error\",\n",
    "              metrics=[\"accuracy\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [],
   "source": [
    "model.compile(optimizer=keras.optimizers.RMSprop(),\n",
    "              loss=keras.losses.MeanSquaredError(),\n",
    "              metrics=[keras.metrics.BinaryAccuracy()])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "### Picking a loss function"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "### Understanding the fit() method"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Calling `fit()` with NumPy data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "16/16 [==============================] - 0s 734us/step - loss: 17.9110 - binary_accuracy: 0.5340\n",
      "Epoch 2/5\n",
      "16/16 [==============================] - 0s 700us/step - loss: 17.2365 - binary_accuracy: 0.5355\n",
      "Epoch 3/5\n",
      "16/16 [==============================] - 0s 667us/step - loss: 16.6602 - binary_accuracy: 0.5360\n",
      "Epoch 4/5\n",
      "16/16 [==============================] - 0s 667us/step - loss: 16.1076 - binary_accuracy: 0.5370\n",
      "Epoch 5/5\n",
      "16/16 [==============================] - 0s 633us/step - loss: 15.5622 - binary_accuracy: 0.5380\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(\n",
    "    inputs,\n",
    "    targets,\n",
    "    epochs=5,\n",
    "    batch_size=128\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'loss': [17.9110164642334,\n",
       "  17.236486434936523,\n",
       "  16.660226821899414,\n",
       "  16.107587814331055,\n",
       "  15.562223434448242],\n",
       " 'binary_accuracy': [0.5339999794960022,\n",
       "  0.5354999899864197,\n",
       "  0.5360000133514404,\n",
       "  0.5370000004768372,\n",
       "  0.5379999876022339]}"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "history.history"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "### Monitoring loss and metrics on validation data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "**Using the `validation_data` argument**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "88/88 [==============================] - 0s 2ms/step - loss: 0.5255 - binary_accuracy: 0.8643 - val_loss: 0.0788 - val_binary_accuracy: 0.9567\n",
      "Epoch 2/5\n",
      "88/88 [==============================] - 0s 1ms/step - loss: 0.0708 - binary_accuracy: 0.9543 - val_loss: 0.0553 - val_binary_accuracy: 0.9750\n",
      "Epoch 3/5\n",
      "88/88 [==============================] - 0s 1ms/step - loss: 0.0754 - binary_accuracy: 0.9600 - val_loss: 0.0564 - val_binary_accuracy: 0.9717\n",
      "Epoch 4/5\n",
      "88/88 [==============================] - 0s 1ms/step - loss: 0.0708 - binary_accuracy: 0.9571 - val_loss: 0.0769 - val_binary_accuracy: 0.9550\n",
      "Epoch 5/5\n",
      "88/88 [==============================] - 0s 1ms/step - loss: 0.0691 - binary_accuracy: 0.9664 - val_loss: 0.0655 - val_binary_accuracy: 0.9650\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x22545186cd0>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = keras.Sequential([keras.layers.Dense(1)])\n",
    "model.compile(optimizer=keras.optimizers.RMSprop(learning_rate=0.1),\n",
    "              loss=keras.losses.MeanSquaredError(),\n",
    "              metrics=[keras.metrics.BinaryAccuracy()])\n",
    "\n",
    "indices_permutation = np.random.permutation(len(inputs))\n",
    "shuffled_inputs = inputs[indices_permutation]\n",
    "shuffled_targets = targets[indices_permutation]\n",
    "\n",
    "num_validation_samples = int(0.3 * len(inputs))\n",
    "val_inputs = shuffled_inputs[:num_validation_samples]\n",
    "val_targets = shuffled_targets[:num_validation_samples]\n",
    "training_inputs = shuffled_inputs[num_validation_samples:]\n",
    "training_targets = shuffled_targets[num_validation_samples:]\n",
    "model.fit(\n",
    "    training_inputs,\n",
    "    training_targets,\n",
    "    epochs=5,\n",
    "    batch_size=16,\n",
    "    validation_data=(val_inputs, val_targets)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "### Inference: Using a model after training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "colab_type": "code"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 0.6424415 ]\n",
      " [ 0.82190853]\n",
      " [ 0.7659918 ]\n",
      " [-0.00859067]\n",
      " [ 0.6902991 ]\n",
      " [ 0.03919756]\n",
      " [-0.17996508]\n",
      " [-0.01434454]\n",
      " [-0.22086555]\n",
      " [ 0.57554984]]\n"
     ]
    }
   ],
   "source": [
    "predictions = model.predict(val_inputs, batch_size=128)\n",
    "print(predictions[:10])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text"
   },
   "source": [
    "## Summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "chapter03_introduction-to-keras-and-tf.i",
   "private_outputs": false,
   "provenance": [],
   "toc_visible": true
  },
  "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}