Lecture4_NN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "870cde38",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## MATH 6388: Statistical and Machine Learning\n",
    "\n",
    "### Lecture 4: Dense/convolutional neural networks, activation functions, and gradient descent\n",
    "\n",
    "### Instructor: Farhad Pourkamali \n",
    "\n",
    "#### Department of Mathematical and Statistical Sciences\n",
    "#### University of Colorado Denver\n",
    "\n",
    "\n",
    "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/farhad-pourkamali/MATH6388/blob/main/Lecture4_NN.ipynb)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3373754",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### Overview \n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* So far, we have seen two linear models for regression and classification problems\n",
    "    * Linear regression: $p(y|\\mathbf{x},\\mathbf{w})=\\mathcal{N}(\\mathbf{w}^T\\mathbf{x}, \\sigma^2)$, where $\\sigma^2$ is a fixed variance for all inputs\n",
    "    * Logistic regression: $p(y|\\mathbf{x},\\mathbf{w})=\\text{Ber}(y|\\sigma(\\mathbf{w}^T\\mathbf{x}))$, where $\\sigma$ is the sigmoid or logistic function\n",
    "\n",
    "* These models make the strong assumption that the input-output mapping is linear \n",
    "    * A better idea: we can equip the feature extractor with its own \"trainable\" parameters\n",
    "    \n",
    "    $$\\mathbf{w}^T\\phi(\\mathbf{x};\\boldsymbol{\\theta})$$ \n",
    "    * Illustrative example in the next slide! "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "953cf35e",
   "metadata": {},
   "source": [
    "### Example, part 1: feature extractor\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* Consider a classification problem with 3 input features $x_1,x_2,x_3$\n",
    "\n",
    "\n",
    "<img src=\"https://github.com/farhad-pourkamali/MATH6388/blob/main/images/nn_example1.png?raw=true\\\" width=700>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95aaa069",
   "metadata": {},
   "source": [
    "### Example, part 2: classifier\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* This is the key idea behind multilayer perceptron (MLP) for \"structured\" or \"tabular\" data $\\mathbf{x}\\in\\mathbb{R}^D$\n",
    "\n",
    "\n",
    "<img src=\"https://github.com/farhad-pourkamali/MATH6388/blob/main/images/nn_example3.png?raw=true\\\" width=600>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb5092f9",
   "metadata": {},
   "source": [
    "### Feedforward networks\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* The information ﬂows through the function being evaluated from $\\mathbf{x}$, through the intermediate computations, and ﬁnally to the output\n",
    "\n",
    "* For example, we might have  $f^{(1)}, f^{(2)}, f^{(3)}$ connected in a chain to form\n",
    "\n",
    "$$f(\\mathbf{x})=f^{(3)}\\Big(f^{(2)}\\big(f^{(1)}(\\mathbf{x})\\big)\\Big)$$\n",
    "\n",
    "* $f^{(1)}$: the first layer,  $f^{(2)}$: the second layer of the network \n",
    "    * Because the training data does not show the desired output for  these layers, they are called **hidden layers**\n",
    "\n",
    "\n",
    "* The overall length of the chain gives the **depth**, and the dimensionality of hidden layers determines the **width** of the model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f25745ca",
   "metadata": {},
   "source": [
    "### Classical example: The XOR problem\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* Learn a function that computes the exclusive OR of its two binary outputs (4 data points in $\\mathbb{R}^2$) \n",
    "    * First activation function: $g(z)=\\text{ReLU}(z)=\\max\\{0,z\\}$ and no activation function for the last layer\n",
    "    \n",
    "<img src=\"https://github.com/farhad-pourkamali/MATH6388/blob/main/images/xor_part1.png?raw=true\\\" width=600>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "325ea0cb",
   "metadata": {},
   "source": [
    "### Classical example: The XOR problem\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* This network correctly predicts all four class labels\n",
    "\n",
    "<img src=\"https://github.com/farhad-pourkamali/MATH6388/blob/main/images/xor_part2.png?raw=true\\\" width=600>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4cf7a1bb",
   "metadata": {},
   "source": [
    "### Why nonlinear activation functions? \n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* If we just use a linear activation function, then the whole model reduces to a regular linear model\n",
    "\n",
    "$$f(\\mathbf{x})=\\mathbf{W}^{(L)}\\ldots\\mathbf{W}^{(2)}\\mathbf{W}^{(1)}\\mathbf{x}=\\mathbf{W}\\mathbf{x}$$\n",
    "\n",
    "* Therefore, it is important to use nonlinear activation functions\n",
    "    * Sigmoid (logistic) function: $\\sigma(a)=\\frac{1}{1+\\exp(-a)}$\n",
    "    * rectified linear unit or ReLU: $\\text{ReLU}(a)=\\max\\{0, a\\}$\n",
    "    * ..."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "528aaa0e",
   "metadata": {},
   "source": [
    "https://www.tensorflow.org/api_docs/python/tf/keras/activations/linear"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ee55aa2",
   "metadata": {},
   "source": [
    "### A tour of activation functions\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* ReLU (Rectified Linear Unit)\n",
    "    * ReLU is known for its simplicity and has been widely used in deep learning due to its ability to introduce non-linearity into the network\n",
    "    \n",
    "    $$\\text{ReLU}(x)=\\max\\{0, x\\}$$\n",
    "    \n",
    "* ELU (Exponential Linear Unit)\n",
    "    * ELU addresses the \"dying ReLU\" problem by allowing negative values, which helps to keep gradients flowing during training\n",
    "    * $\\alpha$ is a hyperparameter that controls the output of the function for negative values of $x$\n",
    "    \n",
    "    $$\\text{ELU}(x)=\\begin{cases}x & \\text{if } x \\geq 0 \\\\\n",
    "\\alpha \\cdot \\big(\\exp(x) - 1\\big) & \\text{if } x < 0\n",
    "\\end{cases}$$\n",
    "\n",
    "* Softplus\n",
    "    * Softplus is a smooth approximation of the ReLU function\n",
    "    * It has the advantage of being differentiable everywhere, which makes it suitable for gradient-based optimization algorithms\n",
    "    \n",
    "    $$\\text{Softplus}(x)=\\log\\big(1+\\exp(x)\\big)$$\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6bec6574",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x300 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Define a range of x values\n",
    "x = np.linspace(-5, 5, 200)  # Adjust the range as needed\n",
    "\n",
    "# Define the ReLU function\n",
    "def relu(x):\n",
    "    return np.maximum(0, x)\n",
    "\n",
    "# Define the ELU function\n",
    "def elu(x, alpha=2.0):\n",
    "    return np.where(x >= 0, x, alpha * (np.exp(x) - 1))\n",
    "\n",
    "# Define the Softplus function\n",
    "def softplus(x):\n",
    "    return np.log(1 + np.exp(x))\n",
    "\n",
    "\n",
    "# Plot the activation functions\n",
    "plt.figure(figsize=(8, 3))\n",
    "\n",
    "plt.subplot(1, 3, 1)\n",
    "plt.title(\"ReLU\")\n",
    "plt.plot(x, relu(x), 'r-')\n",
    "plt.grid()\n",
    "\n",
    "plt.subplot(1, 3, 2)\n",
    "plt.title(r\"ELU ($\\alpha=2$)\")\n",
    "plt.plot(x, elu(x))\n",
    "plt.plot(x, relu(x), 'r:')\n",
    "plt.grid()\n",
    "\n",
    "plt.subplot(1, 3, 3)\n",
    "plt.title(\"Softplus\")\n",
    "plt.plot(x, softplus(x))\n",
    "plt.plot(x, relu(x), 'r:')\n",
    "plt.grid()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6589d788",
   "metadata": {},
   "source": [
    "### Gradient-based learning\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* Neural networks are usually trained by using iterative, gradient-based optimizers such as (stochastic) gradient descent\n",
    "\n",
    "$$\\boldsymbol{\\theta}_{t+1}=\\boldsymbol{\\theta}_{t} - \\eta_t \\nabla J(\\boldsymbol{\\theta}_{t})$$\n",
    "\n",
    "* $\\boldsymbol{\\theta}_{t}$ is the parameter vector at iteration $t$\n",
    "* $\\eta_t$ is the learning rate, determining the step size\n",
    "* $g_t=\\nabla J(\\boldsymbol{\\theta}_{t})$ is the gradient of the loss function \n",
    "\n",
    "* Momentum\n",
    "    * Introduces a velocity term $v_t$ that keeps track of the historical gradients and helps the optimization algorithm navigate through flat or noisy regions\n",
    "    \n",
    "    $$v_{t+1}=\\beta v_{t} + g_t\\\\ \\boldsymbol{\\theta}_{t+1}=\\boldsymbol{\\theta}_{t} - \\eta_t v_{t+1}$$\n",
    "    \n",
    "    * The velocity terms acts like an exponentially weighted moving average of the past gradients\n",
    "    \n",
    "    $$v_{t+1}=\\beta v_{t} + g_t=\\beta \\big(\\beta v_{t-1} + g_{t-1}\\big) + g_t=\\beta^2 v_{t-1} + \\beta g_{t-1} + g_t$$\n",
    "    \n",
    "    $$v_{t+1}=\\beta^3 v_{t-2} + \\beta^2 g_{t-2} + \\beta g_{t-1} + g_t$$\n",
    "    \n",
    "    * The \"scaling factor\" is a geometric series, whose infinite sum is given by ($\\beta<1$)\n",
    "    \n",
    "    $$1+\\beta+\\beta^2+\\ldots = \\sum_{i=0}^\\infty \\beta^i = \\frac{1}{1-\\beta}$$\n",
    "    \n",
    "* Nesterov Momentum \n",
    "    * Tweaks this approach by introducing a \"lookahead step\" before calculating the gradient\n",
    "    * Estimates where the parameter vector will be in the next iteration and then calculates the gradient at that estimated position\n",
    "    * This helps reduce oscillations and accelerates convergence during training, making it a popular choice in deep learning\n",
    "    \n",
    "    $$v_{t+1}=\\beta v_t + \\eta_t \\nabla J(\\boldsymbol{\\theta}_t + \\beta v_t)\\\\ \\boldsymbol{\\theta}_{t+1} = \\boldsymbol{\\theta}_t - v_{t+1}$$\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11c34170",
   "metadata": {},
   "source": [
    "### Backprop\n",
    "\n",
    "* The back-propagation algorithm allows the information from the cost to ﬂow *backward* through the network in order to compute the gradient\n",
    "    * Utilize the chain rule of calculus to compute the derivatives of functions formed by composing other functions\n",
    "    * In the next slide, we look at a simplified model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6194836",
   "metadata": {},
   "source": [
    "### Backprop and chain rule\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* Consider a 3-layer network with one unit in each layer and a scalar input \n",
    "\n",
    "<img src=\"https://github.com/farhad-pourkamali/MATH6388/blob/main/images/backprop1.png?raw=true\\\" width=700>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2b833b87",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Metal device set to: Apple M2 Max\n",
      "df/dx: 11.0\n",
      "df/dy: 5.0\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "# Define the function f(u, v)\n",
    "def f(u, v):\n",
    "    return 3 * u + 5 * v\n",
    "\n",
    "# Define values for x and y\n",
    "x = tf.constant(2.0)\n",
    "y = tf.constant(3.0)\n",
    "\n",
    "# Use tf.GradientTape to compute gradients\n",
    "with tf.GradientTape(persistent=True) as tape:\n",
    "    # Watch the variables we want to compute gradients with respect to\n",
    "    tape.watch(x)\n",
    "    tape.watch(y)\n",
    "    \n",
    "    # Define u and v as functions of x and y\n",
    "    u = 2 * x\n",
    "    v = x + y\n",
    "\n",
    "    # Define f as a function of u and v\n",
    "    result = f(u, v)\n",
    "\n",
    "# Compute the gradients\n",
    "df_dx = tape.gradient(result, x)\n",
    "df_dy = tape.gradient(result, y)\n",
    "\n",
    "\n",
    "# Print the gradients\n",
    "print(\"df/dx:\", df_dx.numpy())\n",
    "print(\"df/dy:\", df_dy.numpy())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf8d6b80",
   "metadata": {},
   "source": [
    "### Implementing MLPs with Keras\n",
    "\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* Keras is a high-level Deep Learning API to build, train, and evaluate all sorts of neural networks\n",
    "    * Keras runs on top of TensorFlow 2\n",
    "    * The best way to learn Keras: https://keras.io/\n",
    "        * Layers \n",
    "        * Models: groups layers into an object with training/inference features \n",
    "        * Optimizers\n",
    "        * Losses\n",
    "        * Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "feb11599",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.12.0\n"
     ]
    }
   ],
   "source": [
    "import numpy as np \n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt \n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "\n",
    "print(tf.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c1e493bd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Example 0 \n",
    "from sklearn.datasets import make_blobs\n",
    "\n",
    "# Set random seed for reproducibility\n",
    "np.random.seed(42)\n",
    "\n",
    "# Create synthetic data set\n",
    "X, y = make_blobs(n_samples=400, centers=[[0, 0], [0, 1], [1, 0], [1, 1]], cluster_std=.1, random_state=42)\n",
    "\n",
    "y[y == 0] = 0\n",
    "y[y == 1] = 1\n",
    "y[y == 2] = 1\n",
    "y[y == 3] = 0\n",
    "\n",
    "\n",
    "# Plot the synthetic data\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, cmap='viridis', edgecolors='k')\n",
    "plt.title('Synthetic Data Set with 2 Classes and 2 Clusters Each')\n",
    "plt.xlabel('Feature 1')\n",
    "plt.ylabel('Feature 2')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "7a9b6215",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 2)                 6         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 1)                 3         \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 9\n",
      "Trainable params: 9\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense\n",
    "\n",
    "# Define the model\n",
    "model = Sequential()\n",
    "\n",
    "# Add the input layer and the hidden layer\n",
    "model.add(Dense(units=2, activation='relu', input_dim=2))  # Assuming 2 input features\n",
    "\n",
    "# Add the output layer\n",
    "model.add(Dense(units=1, activation='sigmoid'))  # Assuming binary classification\n",
    "\n",
    "# Compile the model\n",
    "model.compile(optimizer=tf.keras.optimizers.SGD(learning_rate=0.1), \n",
    "              loss='binary_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "# Display the model summary\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c0d33b99",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([[-0.98041373, -0.29312325],\n",
       "        [ 0.74307454,  0.9272033 ]], dtype=float32),\n",
       " array([0., 0.], dtype=float32)]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.layers[0].get_weights()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "6c0ba519",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([[-0.9563804 ],\n",
       "        [ 0.03683698]], dtype=float32),\n",
       " array([0.], dtype=float32)]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.layers[1].get_weights()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c2676a40",
   "metadata": {},
   "outputs": [
    {
     "ename": "IndexError",
     "evalue": "list index out of range",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[8], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mmodel\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mlayers\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m2\u001b[39;49m\u001b[43m]\u001b[49m\u001b[38;5;241m.\u001b[39mget_weights()\n",
      "\u001b[0;31mIndexError\u001b[0m: list index out of range"
     ]
    }
   ],
   "source": [
    "model.layers[2].get_weights()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c90d9b44",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<keras.layers.core.dense.Dense at 0x1054099c0>,\n",
       " <keras.layers.core.dense.Dense at 0x2de0e37c0>]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "fd4c92c4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      " 1/25 [>.............................] - ETA: 4s - loss: 0.7493 - accuracy: 0.2500"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-10-18 19:58:08.141995: W tensorflow/tsl/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25/25 [==============================] - 0s 3ms/step - loss: 0.7446 - accuracy: 0.2900\n",
      "Epoch 2/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.6941 - accuracy: 0.4775\n",
      "Epoch 3/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.6835 - accuracy: 0.4950\n",
      "Epoch 4/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.6735 - accuracy: 0.5075\n",
      "Epoch 5/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.6586 - accuracy: 0.6775\n",
      "Epoch 6/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.6422 - accuracy: 0.6825\n",
      "Epoch 7/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.6258 - accuracy: 0.7025\n",
      "Epoch 8/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.6088 - accuracy: 0.7250\n",
      "Epoch 9/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5931 - accuracy: 0.7225\n",
      "Epoch 10/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5789 - accuracy: 0.7250\n",
      "Epoch 11/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5651 - accuracy: 0.7300\n",
      "Epoch 12/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5537 - accuracy: 0.7275\n",
      "Epoch 13/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5442 - accuracy: 0.7325\n",
      "Epoch 14/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5361 - accuracy: 0.7325\n",
      "Epoch 15/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5288 - accuracy: 0.7350\n",
      "Epoch 16/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5234 - accuracy: 0.7425\n",
      "Epoch 17/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5187 - accuracy: 0.7425\n",
      "Epoch 18/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5145 - accuracy: 0.7375\n",
      "Epoch 19/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5112 - accuracy: 0.7400\n",
      "Epoch 20/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5084 - accuracy: 0.7400\n",
      "Epoch 21/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5061 - accuracy: 0.7425\n",
      "Epoch 22/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5040 - accuracy: 0.7425\n",
      "Epoch 23/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5023 - accuracy: 0.7450\n",
      "Epoch 24/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.5009 - accuracy: 0.7425\n",
      "Epoch 25/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4992 - accuracy: 0.7475\n",
      "Epoch 26/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4981 - accuracy: 0.7425\n",
      "Epoch 27/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4970 - accuracy: 0.7450\n",
      "Epoch 28/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4959 - accuracy: 0.7475\n",
      "Epoch 29/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4953 - accuracy: 0.7475\n",
      "Epoch 30/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4944 - accuracy: 0.7475\n",
      "Epoch 31/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4936 - accuracy: 0.7475\n",
      "Epoch 32/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4927 - accuracy: 0.7475\n",
      "Epoch 33/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4920 - accuracy: 0.7475\n",
      "Epoch 34/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4916 - accuracy: 0.7475\n",
      "Epoch 35/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4910 - accuracy: 0.7475\n",
      "Epoch 36/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4904 - accuracy: 0.7475\n",
      "Epoch 37/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4899 - accuracy: 0.7475\n",
      "Epoch 38/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4898 - accuracy: 0.7475\n",
      "Epoch 39/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4892 - accuracy: 0.7475\n",
      "Epoch 40/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4887 - accuracy: 0.7475\n",
      "Epoch 41/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4884 - accuracy: 0.7475\n",
      "Epoch 42/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4879 - accuracy: 0.7475\n",
      "Epoch 43/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4877 - accuracy: 0.7475\n",
      "Epoch 44/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4874 - accuracy: 0.7475\n",
      "Epoch 45/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4871 - accuracy: 0.7475\n",
      "Epoch 46/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4868 - accuracy: 0.7475\n",
      "Epoch 47/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4867 - accuracy: 0.7475\n",
      "Epoch 48/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4863 - accuracy: 0.7475\n",
      "Epoch 49/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4860 - accuracy: 0.7475\n",
      "Epoch 50/50\n",
      "25/25 [==============================] - 0s 3ms/step - loss: 0.4860 - accuracy: 0.7475\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(X, y, epochs=50, batch_size=16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "cf1a182c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (6,5)\n",
    "\n",
    "fig, axs = plt.subplots(2)\n",
    "axs[0].plot(history.history['accuracy'])\n",
    "axs[0].set_title('Model accuracy')\n",
    "axs[0].set_xlabel('Epoch')\n",
    "axs[0].set_ylabel('Accuracy')\n",
    "axs[0].set_xlim([0, 49])\n",
    "\n",
    "axs[1].plot(history.history['loss'])\n",
    "axs[1].set_title('Model loss')\n",
    "axs[1].set_xlabel('Epoch')\n",
    "axs[1].set_ylabel('Loss')\n",
    "axs[1].set_xlim([0, 49])\n",
    "\n",
    "plt.tight_layout()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d3c23f5b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([[-0.9665407 , -2.3868065 ],\n",
       "        [ 0.23026928,  2.1624124 ]], dtype=float32),\n",
       " array([-0.45013776, -0.5504174 ], dtype=float32)]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.layers[0].get_weights()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "201aa0e8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([[-0.76309115],\n",
       "        [ 3.1089563 ]], dtype=float32),\n",
       " array([-0.65579224], dtype=float32)]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.layers[1].get_weights()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "f488314e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 37ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualize the classifier \n",
    "\n",
    "plt.rcParams.update({'font.size': 14, \"figure.figsize\": (5,4)})\n",
    "# Create a mesh grid for the entire feature space\n",
    "x_min, x_max = -0.5, 1.5\n",
    "y_min, y_max = -0.5, 1.5\n",
    "step = 0.01\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, step), \n",
    "                     np.arange(y_min, y_max, step))\n",
    "\n",
    "# Flatten the meshgrid and make predictions\n",
    "mesh_input = np.c_[xx.ravel(), yy.ravel()]\n",
    "predictions = model.predict(mesh_input, batch_size=len(mesh_input))\n",
    "\n",
    "# Reshape predictions to the shape of the meshgrid\n",
    "predictions = predictions.reshape(xx.shape)\n",
    "\n",
    "# Create a contour plot\n",
    "plt.contourf(xx, yy, predictions, cmap='viridis', levels=20, alpha=0.8)\n",
    "\n",
    "# Scatter plot of the synthetic data points\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, cmap='viridis', edgecolors='k')\n",
    "\n",
    "plt.title('Contour Plot with Predicted Probabilities')\n",
    "plt.xlabel('Feature 1')\n",
    "plt.ylabel('Feature 2')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f5d7b05e",
   "metadata": {},
   "source": [
    "### Q1: What happens if we don't use the \"sigmoid\" activation function?\n",
    "\n",
    "- https://keras.io/api/layers/core_layers/dense/\n",
    "- https://keras.io/api/losses/probabilistic_losses/\n",
    "\n",
    "When `from_logits=True`, the binary crossentropy loss function in Keras will internally apply the sigmoid activation function during the computation of the loss. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "1396fa81",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "25/25 [==============================] - 0s 5ms/step - loss: 0.7281 - accuracy: 0.5000\n",
      "Epoch 2/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.6770 - accuracy: 0.5000\n",
      "Epoch 3/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.6624 - accuracy: 0.5000\n",
      "Epoch 4/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.6491 - accuracy: 0.5000\n",
      "Epoch 5/50\n",
      "25/25 [==============================] - 0s 5ms/step - loss: 0.6331 - accuracy: 0.5000\n",
      "Epoch 6/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.6151 - accuracy: 0.5000\n",
      "Epoch 7/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.5945 - accuracy: 0.5000\n",
      "Epoch 8/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.5699 - accuracy: 0.5125\n",
      "Epoch 9/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.5430 - accuracy: 0.5375\n",
      "Epoch 10/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.5099 - accuracy: 0.6175\n",
      "Epoch 11/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.4734 - accuracy: 0.9875\n",
      "Epoch 12/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.4331 - accuracy: 0.9725\n",
      "Epoch 13/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.3907 - accuracy: 0.9925\n",
      "Epoch 14/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.3477 - accuracy: 0.9950\n",
      "Epoch 15/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.3085 - accuracy: 1.0000\n",
      "Epoch 16/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.2693 - accuracy: 0.9950\n",
      "Epoch 17/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.2356 - accuracy: 1.0000\n",
      "Epoch 18/50\n",
      "25/25 [==============================] - 0s 5ms/step - loss: 0.2052 - accuracy: 1.0000\n",
      "Epoch 19/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.1802 - accuracy: 1.0000\n",
      "Epoch 20/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.1585 - accuracy: 1.0000\n",
      "Epoch 21/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.1405 - accuracy: 1.0000\n",
      "Epoch 22/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.1252 - accuracy: 1.0000\n",
      "Epoch 23/50\n",
      "25/25 [==============================] - 0s 5ms/step - loss: 0.1120 - accuracy: 1.0000\n",
      "Epoch 24/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.1011 - accuracy: 1.0000\n",
      "Epoch 25/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0917 - accuracy: 1.0000\n",
      "Epoch 26/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0838 - accuracy: 1.0000\n",
      "Epoch 27/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0767 - accuracy: 1.0000\n",
      "Epoch 28/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0710 - accuracy: 1.0000\n",
      "Epoch 29/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0657 - accuracy: 1.0000\n",
      "Epoch 30/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0610 - accuracy: 1.0000\n",
      "Epoch 31/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0568 - accuracy: 1.0000\n",
      "Epoch 32/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0533 - accuracy: 1.0000\n",
      "Epoch 33/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0500 - accuracy: 1.0000\n",
      "Epoch 34/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0472 - accuracy: 1.0000\n",
      "Epoch 35/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0446 - accuracy: 1.0000\n",
      "Epoch 36/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0421 - accuracy: 1.0000\n",
      "Epoch 37/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0401 - accuracy: 1.0000\n",
      "Epoch 38/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0382 - accuracy: 1.0000\n",
      "Epoch 39/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0363 - accuracy: 1.0000\n",
      "Epoch 40/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0346 - accuracy: 1.0000\n",
      "Epoch 41/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0331 - accuracy: 1.0000\n",
      "Epoch 42/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0318 - accuracy: 1.0000\n",
      "Epoch 43/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0304 - accuracy: 1.0000\n",
      "Epoch 44/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0293 - accuracy: 1.0000\n",
      "Epoch 45/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0281 - accuracy: 1.0000\n",
      "Epoch 46/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0271 - accuracy: 1.0000\n",
      "Epoch 47/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0262 - accuracy: 1.0000\n",
      "Epoch 48/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0252 - accuracy: 1.0000\n",
      "Epoch 49/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0244 - accuracy: 1.0000\n",
      "Epoch 50/50\n",
      "25/25 [==============================] - 0s 4ms/step - loss: 0.0236 - accuracy: 1.0000\n"
     ]
    }
   ],
   "source": [
    "# Define the model\n",
    "model = Sequential()\n",
    "\n",
    "# Add the input layer and the hidden layer\n",
    "model.add(Dense(units=2, activation='elu', input_dim=2))  \n",
    "\n",
    "# Add the output layer\n",
    "model.add(Dense(units=1)) # no 'sigmoid' activation \n",
    "\n",
    "model.compile(optimizer=tf.keras.optimizers.SGD(learning_rate=0.1), \n",
    "              loss=tf.keras.losses.BinaryCrossentropy(from_logits=True), metrics=['accuracy'])\n",
    "\n",
    "history = model.fit(X, y, epochs=50, batch_size=16)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "61b8b6e9",
   "metadata": {},
   "source": [
    "### Q2: How to set aside validation data? \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "4f72a446",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "23/23 [==============================] - 0s 13ms/step - loss: 0.7590 - accuracy: 0.5028 - val_loss: 0.7186 - val_accuracy: 0.4750\n",
      "Epoch 2/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.7178 - accuracy: 0.5000 - val_loss: 0.7023 - val_accuracy: 0.4750\n",
      "Epoch 3/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.7073 - accuracy: 0.5028 - val_loss: 0.6956 - val_accuracy: 0.4750\n",
      "Epoch 4/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.7013 - accuracy: 0.5028 - val_loss: 0.6915 - val_accuracy: 0.4750\n",
      "Epoch 5/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6970 - accuracy: 0.5028 - val_loss: 0.6892 - val_accuracy: 0.4750\n",
      "Epoch 6/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6938 - accuracy: 0.5028 - val_loss: 0.6863 - val_accuracy: 0.4750\n",
      "Epoch 7/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6918 - accuracy: 0.5028 - val_loss: 0.6850 - val_accuracy: 0.4750\n",
      "Epoch 8/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6891 - accuracy: 0.5028 - val_loss: 0.6830 - val_accuracy: 0.4750\n",
      "Epoch 9/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6881 - accuracy: 0.5028 - val_loss: 0.6816 - val_accuracy: 0.4750\n",
      "Epoch 10/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6868 - accuracy: 0.5028 - val_loss: 0.6806 - val_accuracy: 0.4750\n",
      "Epoch 11/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6851 - accuracy: 0.5028 - val_loss: 0.6804 - val_accuracy: 0.4750\n",
      "Epoch 12/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6838 - accuracy: 0.5028 - val_loss: 0.6797 - val_accuracy: 0.4750\n",
      "Epoch 13/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6824 - accuracy: 0.5028 - val_loss: 0.6774 - val_accuracy: 0.4750\n",
      "Epoch 14/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6802 - accuracy: 0.5028 - val_loss: 0.6755 - val_accuracy: 0.4750\n",
      "Epoch 15/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6782 - accuracy: 0.5028 - val_loss: 0.6747 - val_accuracy: 0.4750\n",
      "Epoch 16/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6768 - accuracy: 0.5028 - val_loss: 0.6732 - val_accuracy: 0.4750\n",
      "Epoch 17/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6741 - accuracy: 0.5028 - val_loss: 0.6700 - val_accuracy: 0.4750\n",
      "Epoch 18/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6718 - accuracy: 0.5028 - val_loss: 0.6679 - val_accuracy: 0.4750\n",
      "Epoch 19/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6689 - accuracy: 0.5028 - val_loss: 0.6639 - val_accuracy: 0.5000\n",
      "Epoch 20/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6658 - accuracy: 0.5278 - val_loss: 0.6620 - val_accuracy: 0.5000\n",
      "Epoch 21/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6617 - accuracy: 0.5028 - val_loss: 0.6578 - val_accuracy: 0.5000\n",
      "Epoch 22/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6579 - accuracy: 0.5389 - val_loss: 0.6543 - val_accuracy: 0.5250\n",
      "Epoch 23/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6530 - accuracy: 0.6278 - val_loss: 0.6517 - val_accuracy: 0.5250\n",
      "Epoch 24/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6490 - accuracy: 0.6083 - val_loss: 0.6461 - val_accuracy: 0.6250\n",
      "Epoch 25/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6437 - accuracy: 0.6861 - val_loss: 0.6420 - val_accuracy: 0.6500\n",
      "Epoch 26/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6379 - accuracy: 0.6639 - val_loss: 0.6343 - val_accuracy: 0.7000\n",
      "Epoch 27/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6323 - accuracy: 0.7361 - val_loss: 0.6288 - val_accuracy: 0.7000\n",
      "Epoch 28/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6256 - accuracy: 0.7444 - val_loss: 0.6235 - val_accuracy: 0.7000\n",
      "Epoch 29/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6194 - accuracy: 0.7389 - val_loss: 0.6162 - val_accuracy: 0.7250\n",
      "Epoch 30/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6134 - accuracy: 0.7500 - val_loss: 0.6094 - val_accuracy: 0.7250\n",
      "Epoch 31/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.6068 - accuracy: 0.7500 - val_loss: 0.6030 - val_accuracy: 0.7250\n",
      "Epoch 32/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5993 - accuracy: 0.7528 - val_loss: 0.5955 - val_accuracy: 0.7250\n",
      "Epoch 33/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5913 - accuracy: 0.7528 - val_loss: 0.5887 - val_accuracy: 0.7250\n",
      "Epoch 34/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5848 - accuracy: 0.7528 - val_loss: 0.5819 - val_accuracy: 0.7250\n",
      "Epoch 35/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5781 - accuracy: 0.7528 - val_loss: 0.5737 - val_accuracy: 0.7250\n",
      "Epoch 36/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5700 - accuracy: 0.7528 - val_loss: 0.5666 - val_accuracy: 0.7250\n",
      "Epoch 37/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5640 - accuracy: 0.7500 - val_loss: 0.5602 - val_accuracy: 0.7250\n",
      "Epoch 38/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5568 - accuracy: 0.7528 - val_loss: 0.5532 - val_accuracy: 0.7250\n",
      "Epoch 39/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5495 - accuracy: 0.7528 - val_loss: 0.5467 - val_accuracy: 0.7250\n",
      "Epoch 40/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5435 - accuracy: 0.7528 - val_loss: 0.5397 - val_accuracy: 0.7250\n",
      "Epoch 41/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5369 - accuracy: 0.7528 - val_loss: 0.5346 - val_accuracy: 0.7250\n",
      "Epoch 42/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5304 - accuracy: 0.7528 - val_loss: 0.5281 - val_accuracy: 0.7250\n",
      "Epoch 43/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5254 - accuracy: 0.7528 - val_loss: 0.5209 - val_accuracy: 0.7250\n",
      "Epoch 44/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5189 - accuracy: 0.7528 - val_loss: 0.5162 - val_accuracy: 0.7250\n",
      "Epoch 45/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5131 - accuracy: 0.7528 - val_loss: 0.5101 - val_accuracy: 0.7250\n",
      "Epoch 46/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5079 - accuracy: 0.7528 - val_loss: 0.5038 - val_accuracy: 0.7250\n",
      "Epoch 47/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.5033 - accuracy: 0.7528 - val_loss: 0.4994 - val_accuracy: 0.7250\n",
      "Epoch 48/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.4985 - accuracy: 0.7528 - val_loss: 0.4951 - val_accuracy: 0.7250\n",
      "Epoch 49/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.4930 - accuracy: 0.7528 - val_loss: 0.4900 - val_accuracy: 0.7250\n",
      "Epoch 50/50\n",
      "23/23 [==============================] - 0s 6ms/step - loss: 0.4887 - accuracy: 0.7528 - val_loss: 0.4854 - val_accuracy: 0.7250\n"
     ]
    }
   ],
   "source": [
    "# Define the model\n",
    "model = Sequential()\n",
    "\n",
    "# Add the input layer and the hidden layer\n",
    "model.add(Dense(units=2, activation='elu', input_dim=2))  \n",
    "\n",
    "# Add the output layer\n",
    "model.add(Dense(units=1)) # no 'sigmoid' activation \n",
    "\n",
    "model.compile(optimizer=tf.keras.optimizers.SGD(learning_rate=0.1), \n",
    "              loss=tf.keras.losses.BinaryCrossentropy(from_logits=True), metrics=['accuracy'])\n",
    "\n",
    "history = model.fit(X, y, epochs=50, batch_size=16, validation_split=0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "3cfa9fb5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "history.history.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "4a394f09",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.DataFrame(history.history).plot(figsize=(5, 3))\n",
    "plt.grid(True)\n",
    "plt.gca().set_ylim(0, 1.1)\n",
    "plt.xlabel('epochs')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "37930233",
   "metadata": {},
   "source": [
    "### Q3: How to ensure reproducible results? \n",
    "\n",
    "- Set random seeds for Python, NumPy, and TensorFlow. This ensures that the random number generation is consistent across runs.\n",
    "- When creating your Keras model, set the initial weights using a fixed seed. This is especially important if your model architecture involves random weight initialization.\n",
    "- HeUniform\n",
    "    - Draws samples from a uniform distribution within [-limit, limit], where limit = sqrt(6 / fan_in) (fan_in is the number of input units in the weight tensor).\n",
    "    - https://keras.io/api/layers/initializers/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "1ebb0bc9",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[array([[3.205899 ],\n",
       "        [3.3351045]], dtype=float32),\n",
       " array([0.15003212], dtype=float32)]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "seed_value = 4\n",
    "\n",
    "tf.random.set_seed(seed_value)\n",
    "\n",
    "# Define the model\n",
    "model = Sequential()\n",
    "\n",
    "# Add the input layer and the hidden layer\n",
    "model.add(Dense(units=2, activation='elu', input_dim=2, \n",
    "                kernel_initializer= tf.keras.initializers.HeUniform(seed=seed_value)))  \n",
    "\n",
    "# Add the output layer\n",
    "model.add(Dense(units=1, \n",
    "          kernel_initializer= tf.keras.initializers.HeUniform(seed=seed_value))) \n",
    "\n",
    "model.compile(optimizer=tf.keras.optimizers.SGD(learning_rate=0.1), \n",
    "              loss=tf.keras.losses.BinaryCrossentropy(from_logits=True), metrics=['accuracy'])\n",
    "\n",
    "history = model.fit(X, y, epochs=50, batch_size=16, validation_split=0.1, verbose=0)\n",
    "\n",
    "model.layers[1].get_weights()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "9a6b1494",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 21ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualize the classifier \n",
    "\n",
    "plt.rcParams.update({'font.size': 14, \"figure.figsize\": (5,4)})\n",
    "# Create a mesh grid for the entire feature space\n",
    "x_min, x_max = -0.5, 1.5\n",
    "y_min, y_max = -0.5, 1.5\n",
    "step = 0.01\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, step), \n",
    "                     np.arange(y_min, y_max, step))\n",
    "\n",
    "# Flatten the meshgrid and make predictions\n",
    "mesh_input = np.c_[xx.ravel(), yy.ravel()]\n",
    "predictions = model.predict(mesh_input, batch_size=len(mesh_input))\n",
    "\n",
    "# Reshape predictions to the shape of the meshgrid\n",
    "predictions = predictions.reshape(xx.shape)\n",
    "\n",
    "# Create a contour plot\n",
    "plt.contourf(xx, yy, predictions, cmap='viridis', levels=20, alpha=0.8)\n",
    "\n",
    "# Scatter plot of the synthetic data points\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y, cmap='viridis', edgecolors='k')\n",
    "\n",
    "plt.title('Contour Plot with Predicted Probabilities')\n",
    "plt.xlabel('Feature 1')\n",
    "plt.ylabel('Feature 2')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ad28798d",
   "metadata": {},
   "source": [
    "### Q4: How to perform hyperparameter optimization?\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "22f8f4c2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(300, 2) (100, 2)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Split the data into training and testing/validation sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y,\n",
    "                                                    test_size=0.25, random_state=42)\n",
    "\n",
    "print(X_train.shape, X_test.shape)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "9e914a12",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_model(num_hidden_layers=1, num_units=32):\n",
    "    \n",
    "    model = Sequential()\n",
    "    \n",
    "    model.add(Dense(num_units, input_dim=X_train.shape[1], activation='elu'))\n",
    "\n",
    "    for _ in range(num_hidden_layers - 1):\n",
    "        \n",
    "        model.add(Dense(num_units, activation='elu'))\n",
    "\n",
    "    model.add(Dense(1, activation='sigmoid'))\n",
    "\n",
    "    model.compile(loss='binary_crossentropy', \n",
    "                  optimizer=tf.keras.optimizers.legacy.SGD(learning_rate=0.1), \n",
    "                  metrics=['accuracy'])\n",
    "    \n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "4eafaa01",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training model with 1 hidden layers and 2 units.\n",
      "4/4 [==============================] - 0s 4ms/step\n",
      "Accuracy: 0.98\n",
      "Training model with 1 hidden layers and 4 units.\n",
      "4/4 [==============================] - 0s 4ms/step\n",
      "Accuracy: 1.0\n",
      "Training model with 1 hidden layers and 6 units.\n",
      "WARNING:tensorflow:5 out of the last 11 calls to <function Model.make_predict_function.<locals>.predict_function at 0x2e059d1b0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:5 out of the last 11 calls to <function Model.make_predict_function.<locals>.predict_function at 0x2e059d1b0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4/4 [==============================] - 0s 4ms/step\n",
      "Accuracy: 1.0\n",
      "Training model with 2 hidden layers and 2 units.\n",
      "4/4 [==============================] - 0s 4ms/step\n",
      "Accuracy: 1.0\n",
      "Training model with 2 hidden layers and 4 units.\n",
      "4/4 [==============================] - 0s 5ms/step\n",
      "Accuracy: 1.0\n",
      "Training model with 2 hidden layers and 6 units.\n",
      "4/4 [==============================] - 0s 5ms/step\n",
      "Accuracy: 1.0\n",
      "Training model with 3 hidden layers and 2 units.\n",
      "4/4 [==============================] - 0s 6ms/step\n",
      "Accuracy: 0.99\n",
      "Training model with 3 hidden layers and 4 units.\n",
      "4/4 [==============================] - 0s 6ms/step\n",
      "Accuracy: 1.0\n",
      "Training model with 3 hidden layers and 6 units.\n",
      "4/4 [==============================] - 0s 6ms/step\n",
      "Accuracy: 1.0\n",
      "Best parameters: {'num_hidden_layers': 1, 'num_units': 4}\n",
      "Best accuracy: 1.0\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "hidden_layers_list = [1, 2, 3]\n",
    "\n",
    "num_units_list = [2, 4, 6]\n",
    "\n",
    "best_accuracy = 0\n",
    "\n",
    "best_params = {}\n",
    "\n",
    "# Grid search\n",
    "for num_hidden_layers in hidden_layers_list:\n",
    "    \n",
    "    for num_units in num_units_list:\n",
    "        \n",
    "        print(f\"Training model with {num_hidden_layers} hidden layers and {num_units} units.\")\n",
    "\n",
    "        model = create_model(num_hidden_layers=num_hidden_layers, num_units=num_units)\n",
    "        \n",
    "        model.fit(X_train, y_train, epochs=50, batch_size=16, verbose=0)\n",
    "\n",
    "        # Evaluate on the test set\n",
    "        y_pred = (model.predict(X_test)>0.5).astype(int).ravel()\n",
    "        accuracy = accuracy_score(y_test, y_pred)\n",
    "\n",
    "        print(f\"Accuracy: {accuracy}\")\n",
    "\n",
    "        # Update best parameters if accuracy is improved\n",
    "        if accuracy > best_accuracy:\n",
    "            best_accuracy = accuracy\n",
    "            best_params = {'num_hidden_layers': num_hidden_layers, 'num_units': num_units}\n",
    "\n",
    "print(\"Best parameters:\", best_params)\n",
    "\n",
    "print(\"Best accuracy:\", best_accuracy)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f8dcd98",
   "metadata": {},
   "source": [
    "### We can add the choice of activation function "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "05714cf8",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_model(num_hidden_layers=1, num_units=32, activation='relu'):\n",
    "    \n",
    "    model = Sequential()\n",
    "    \n",
    "    model.add(Dense(num_units, input_dim=X_train.shape[1], activation=activation))\n",
    "\n",
    "    for _ in range(num_hidden_layers - 1):\n",
    "        model.add(Dense(num_units, activation=activation))\n",
    "\n",
    "    model.add(Dense(1, activation='sigmoid'))\n",
    "\n",
    "    model.compile(loss='binary_crossentropy', optimizer=tf.keras.optimizers.legacy.SGD(learning_rate=0.1),\n",
    "                  metrics=['accuracy'])\n",
    "    \n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "ff3bd788",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training model with 1 hidden layers, 2 units, and activation function relu.\n",
      "4/4 [==============================] - 0s 3ms/step\n",
      "Accuracy: 1.0\n",
      "Training model with 1 hidden layers, 2 units, and activation function elu.\n",
      "4/4 [==============================] - 0s 4ms/step\n",
      "Accuracy: 0.65\n",
      "Training model with 1 hidden layers, 2 units, and activation function swish.\n",
      "4/4 [==============================] - 0s 7ms/step\n",
      "Accuracy: 0.82\n",
      "Training model with 1 hidden layers, 4 units, and activation function relu.\n",
      "4/4 [==============================] - 0s 4ms/step\n",
      "Accuracy: 0.67\n",
      "Training model with 1 hidden layers, 4 units, and activation function elu.\n",
      "4/4 [==============================] - 0s 4ms/step\n",
      "Accuracy: 0.99\n",
      "Training model with 1 hidden layers, 4 units, and activation function swish.\n",
      "4/4 [==============================] - 0s 5ms/step\n",
      "Accuracy: 0.99\n",
      "Training model with 2 hidden layers, 2 units, and activation function relu.\n",
      "4/4 [==============================] - 0s 5ms/step\n",
      "Accuracy: 0.42\n",
      "Training model with 2 hidden layers, 2 units, and activation function elu.\n",
      "4/4 [==============================] - 0s 6ms/step\n",
      "Accuracy: 0.68\n",
      "Training model with 2 hidden layers, 2 units, and activation function swish.\n",
      "4/4 [==============================] - 0s 6ms/step\n",
      "Accuracy: 0.75\n",
      "Training model with 2 hidden layers, 4 units, and activation function relu.\n",
      "4/4 [==============================] - 0s 6ms/step\n",
      "Accuracy: 0.99\n",
      "Training model with 2 hidden layers, 4 units, and activation function elu.\n",
      "4/4 [==============================] - 0s 5ms/step\n",
      "Accuracy: 0.99\n",
      "Training model with 2 hidden layers, 4 units, and activation function swish.\n",
      "4/4 [==============================] - 0s 7ms/step\n",
      "Accuracy: 1.0\n",
      "Training model with 3 hidden layers, 2 units, and activation function relu.\n",
      "4/4 [==============================] - 0s 6ms/step\n",
      "Accuracy: 0.99\n",
      "Training model with 3 hidden layers, 2 units, and activation function elu.\n",
      "4/4 [==============================] - 0s 6ms/step\n",
      "Accuracy: 1.0\n",
      "Training model with 3 hidden layers, 2 units, and activation function swish.\n",
      "4/4 [==============================] - 0s 8ms/step\n",
      "Accuracy: 0.42\n",
      "Training model with 3 hidden layers, 4 units, and activation function relu.\n",
      "4/4 [==============================] - 0s 5ms/step\n",
      "Accuracy: 1.0\n",
      "Training model with 3 hidden layers, 4 units, and activation function elu.\n",
      "4/4 [==============================] - 0s 9ms/step\n",
      "Accuracy: 1.0\n",
      "Training model with 3 hidden layers, 4 units, and activation function swish.\n",
      "4/4 [==============================] - 0s 14ms/step\n",
      "Accuracy: 0.72\n",
      "Best parameters: {'num_hidden_layers': 1, 'num_units': 2, 'activation_function': 'relu'}\n",
      "Best accuracy: 1.0\n"
     ]
    }
   ],
   "source": [
    "hidden_layers_list = [1, 2, 3]\n",
    "num_units_list = [2, 4]\n",
    "activation_functions = ['relu', 'elu', 'swish']\n",
    "\n",
    "best_accuracy = 0\n",
    "best_params = {}\n",
    "\n",
    "# Grid search\n",
    "for num_hidden_layers in hidden_layers_list:\n",
    "    for num_units in num_units_list:\n",
    "        for activation_function in activation_functions:\n",
    "            \n",
    "            print(f\"Training model with {num_hidden_layers} hidden layers, {num_units} units, and activation function {activation_function}.\")\n",
    "\n",
    "            model = create_model(num_hidden_layers=num_hidden_layers, \n",
    "                                 num_units=num_units, activation=activation_function)\n",
    "                  \n",
    "            model.fit(X_train, y_train, epochs=20, batch_size=16, verbose=0)\n",
    "\n",
    "            # Evaluate on the test set\n",
    "            y_pred = (model.predict(X_test)>0.5).astype(int).ravel()\n",
    "            accuracy = accuracy_score(y_test, y_pred)\n",
    "\n",
    "            print(f\"Accuracy: {accuracy}\")\n",
    "\n",
    "            # Update best parameters if accuracy is improved\n",
    "            if accuracy > best_accuracy:\n",
    "                best_accuracy = accuracy\n",
    "                best_params = {'num_hidden_layers': num_hidden_layers, 'num_units': num_units, 'activation_function': activation_function}\n",
    "\n",
    "print(\"Best parameters:\", best_params)\n",
    "print(\"Best accuracy:\", best_accuracy)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e5520987",
   "metadata": {},
   "source": [
    "### First example\n",
    "\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* Building an Image Classifier using the Sequential API\n",
    "    * Work with the Fashion MNIST data set\n",
    "    * Steps: \n",
    "        * Preparing the data\n",
    "        * Creating the model (number of hidden layers, ...)\n",
    "        * Compiling the model (loss function, optimizer, ...)\n",
    "        * Training the model \n",
    "        * Using the model to make predictions \n",
    "        \n",
    "<img src=\"https://github.com/farhad-pourkamali/MATH6388/blob/main/images/fmnist.png?raw=true\\\" width=400>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "20665ab1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(60000, 28, 28) (10000, 28, 28) uint8\n"
     ]
    }
   ],
   "source": [
    "# The dataset is already divided into train/test sets \n",
    "\n",
    "fashion_mnist = keras.datasets.fashion_mnist\n",
    "\n",
    "(X_train_full, y_train_full), (X_test, y_test) = fashion_mnist.load_data()\n",
    "\n",
    "print(X_train_full.shape, X_test.shape, X_train_full.dtype)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "029aabd3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "float64\n"
     ]
    }
   ],
   "source": [
    "# Split the train set into a validation set and a (smaller) train set\n",
    "\n",
    "X_valid, X_train = X_train_full[:5000] / 255., X_train_full[5000:] / 255.\n",
    "y_valid, y_train = y_train_full[:5000], y_train_full[5000:]\n",
    "X_test = X_test / 255.\n",
    "\n",
    "print(X_test.dtype)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "dc968cbb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([4, 0, 7, 9, 9, 9, 4, 4, 3, 4], dtype=uint8)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Type of labels\n",
    "\n",
    "y_train[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "ccfbd362",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype=uint8),\n",
       " array([5543, 5444, 5496, 5499, 5512, 5507, 5507, 5488, 5510, 5494]))"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(y_train, return_counts=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "7913b531",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x240 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(7.2, 2.4))\n",
    "\n",
    "plt.imshow(X_train[3], cmap='binary')\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "4aa9d06f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating the model using the Sequential API\n",
    "\n",
    "model = keras.models.Sequential()\n",
    "\n",
    "model.add(keras.layers.Flatten(input_shape=[28, 28]))\n",
    "model.add(keras.layers.Dense(300, activation=\"relu\")) # 300 neurons, activation function: ReLU \n",
    "model.add(keras.layers.Dense(100, activation=\"relu\")) # 100 neurons\n",
    "model.add(keras.layers.Dense(10, activation=\"softmax\")) # 10 classes so \"softmax\"\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "2a1695fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Another way to create the model with a \"list\" of layers\n",
    "\n",
    "model = keras.models.Sequential([\n",
    "    keras.layers.Flatten(input_shape=[28, 28]),\n",
    "    keras.layers.Dense(300, activation=\"relu\"),\n",
    "    keras.layers.Dense(100, activation=\"relu\"),\n",
    "    keras.layers.Dense(10, activation=\"softmax\")\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "dccd63b9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_32\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " flatten_1 (Flatten)         (None, 784)               0         \n",
      "                                                                 \n",
      " dense_92 (Dense)            (None, 300)               235500    \n",
      "                                                                 \n",
      " dense_93 (Dense)            (None, 100)               30100     \n",
      "                                                                 \n",
      " dense_94 (Dense)            (None, 10)                1010      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 266,610\n",
      "Trainable params: 266,610\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "0d3e6027",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "266610"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# How can we calculate the number of parameters? \n",
    "\n",
    "300*785 + 100 * 301 + 10 * 101"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "97b73008",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Visualize the model \n",
    "\n",
    "from keras.utils import plot_model\n",
    "\n",
    "plot_model(model, to_file='model.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "aa1eb4bc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Name layers \n",
    "model = keras.models.Sequential([\n",
    "    keras.layers.Flatten(input_shape=[28, 28], name='flat'),\n",
    "    keras.layers.Dense(300, activation=\"relu\", name='hidden1'),\n",
    "    keras.layers.Dense(100, activation=\"relu\", name='hidden2'),\n",
    "    keras.layers.Dense(10, activation=\"softmax\", name='output')\n",
    "])\n",
    "\n",
    "plot_model(model, to_file='model.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "b4ecc0da",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<keras.layers.reshaping.flatten.Flatten at 0x30491c850>,\n",
       " <keras.layers.core.dense.Dense at 0x302e71e40>,\n",
       " <keras.layers.core.dense.Dense at 0x302e72fe0>,\n",
       " <keras.layers.core.dense.Dense at 0x302e70490>]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "630831ee",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "hidden1\n"
     ]
    }
   ],
   "source": [
    "# All the parameters of a layer can be accessed using its \n",
    "\n",
    "layer = model.layers[1]\n",
    "\n",
    "print(layer.name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "25bb57e0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(784, 300) (300,)\n"
     ]
    }
   ],
   "source": [
    "weights, biases = layer.get_weights()\n",
    "\n",
    "print(weights.shape, biases.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "a345ecce",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compiling the model \n",
    "\n",
    "model.compile(loss=\"sparse_categorical_crossentropy\", # \"sparse\" because labels are 0,1,2,...\n",
    "              optimizer=\"sgd\",\n",
    "              metrics=[\"accuracy\"]) "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "064b6d3e",
   "metadata": {},
   "source": [
    "What is `sgd`? https://www.tensorflow.org/api_docs/python/tf/keras/optimizers/legacy/SGD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "5fbfc6e7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "860/860 [==============================] - 4s 5ms/step - loss: 0.8432 - accuracy: 0.7311 - val_loss: 0.5852 - val_accuracy: 0.8038\n",
      "Epoch 2/5\n",
      "860/860 [==============================] - 4s 4ms/step - loss: 0.5460 - accuracy: 0.8154 - val_loss: 0.5010 - val_accuracy: 0.8318\n",
      "Epoch 3/5\n",
      "860/860 [==============================] - 4s 4ms/step - loss: 0.4901 - accuracy: 0.8292 - val_loss: 0.4720 - val_accuracy: 0.8404\n",
      "Epoch 4/5\n",
      "860/860 [==============================] - 4s 4ms/step - loss: 0.4606 - accuracy: 0.8384 - val_loss: 0.4542 - val_accuracy: 0.8456\n",
      "Epoch 5/5\n",
      "860/860 [==============================] - 4s 4ms/step - loss: 0.4383 - accuracy: 0.8467 - val_loss: 0.4661 - val_accuracy: 0.8386\n"
     ]
    }
   ],
   "source": [
    "# Training the model \n",
    "\n",
    "history = model.fit(X_train, y_train, batch_size=64, epochs=5,\n",
    "                    validation_data=(X_valid, y_valid))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "d115edeb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "860.0"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Why do we have 860 updates in each epoch? \n",
    "\n",
    "np.ceil(55000/64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "b334d838",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# What is included in history? \n",
    "\n",
    "history.history.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "17c88e8f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.DataFrame(history.history).plot(figsize=(5, 3))\n",
    "plt.grid(True)\n",
    "plt.gca().set_ylim(0, 1)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "e38e9180",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 102ms/step - loss: 0.5060 - accuracy: 0.8236\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.506011426448822, 0.8236000537872314]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# evaluate the model \n",
    "\n",
    "model.evaluate(X_test, y_test, batch_size=10000) # one batch \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "fc5b2f90",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 41ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0.  , 0.  , 0.  , 0.  , 0.  , 0.17, 0.  , 0.42, 0.  , 0.41],\n",
       "       [0.  , 0.  , 1.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  ],\n",
       "       [0.  , 1.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  ]],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# making predictions \n",
    "\n",
    "X_new = X_test[:3]\n",
    "y_proba = model.predict(X_new)\n",
    "y_proba.round(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "fc742609",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([7, 2, 1])"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = np.argmax(y_proba, axis=1)\n",
    "y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "ede42b7d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x300 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "class_names = [\"T-shirt/top\", \"Trouser\", \"Pullover\", \"Dress\", \"Coat\",\n",
    "               \"Sandal\", \"Shirt\", \"Sneaker\", \"Bag\", \"Ankle boot\"]\n",
    "\n",
    "plt.figure(figsize=(6, 3))\n",
    "for index, image in enumerate(X_new):\n",
    "    plt.subplot(1, 3, index + 1)\n",
    "    plt.imshow(image, cmap=\"binary\", interpolation=\"nearest\")\n",
    "    plt.axis('off')\n",
    "    plt.title(class_names[y_pred[index]], fontsize=12)\n",
    "plt.subplots_adjust(wspace=0.2, hspace=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c785e6c0",
   "metadata": {},
   "source": [
    "### Second example\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* In this illustration, we delve into the influence of activation functions on the precision of neural networks, exploring the effects of three distinct activation functions\n",
    "    * Rectified Linear Unit (ReLU)\n",
    "    * Exponential Linear Unit (ELU)\n",
    "    * Softplus\n",
    "\n",
    "* The Functional API in Keras allows you to create custom neural network architectures with more flexibility than the Sequential API\n",
    "    * Define Input Layer: `input_layer = Input(shape=(input_shape,))`\n",
    "    * Create Layers and Connect Them: `hidden_layer = Dense(128, activation='relu')(input_layer)` and `output_layer = Dense(output_shape, activation='softmax')(hidden_layer)`\n",
    "    * Model Declaration: `model = keras.Model(inputs=input_layer, outputs=output_layer)`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "56394eb2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Set random seed for reproducibility\n",
    "np.random.seed(5)\n",
    "\n",
    "# Generate synthetic data for three-class classification\n",
    "mean_class_1 = [2, 4]\n",
    "cov_class_1 = [[1, 0.5], [0.5, 1]]\n",
    "data_class_1 = np.random.multivariate_normal(mean_class_1, cov_class_1, 200)\n",
    "\n",
    "mean_class_2 = [6, 7]\n",
    "cov_class_2 = [[1, -0.5], [-0.5, 1]]\n",
    "data_class_2 = np.random.multivariate_normal(mean_class_2, cov_class_2, 200)\n",
    "\n",
    "mean_class_3 = [2, 7]\n",
    "cov_class_3 = [[1, 0], [0, 1]]\n",
    "data_class_3 = np.random.multivariate_normal(mean_class_3, cov_class_3, 200)\n",
    "\n",
    "# Combine the three classes into a single dataset\n",
    "X = np.vstack((data_class_1, data_class_2, data_class_3))\n",
    "y = np.hstack((np.zeros(200), np.ones(200), 2 * np.ones(200)))  \n",
    "\n",
    "# Plot the synthetic data\n",
    "plt.rcParams.update({'font.size': 11, \"figure.figsize\": (4,3)})\n",
    "colors = {0: \"#1f77b4\", 1: \"#ff7f0e\", 2:  \"#2ca02c\", 3: \"#d62728\"}\n",
    "y_col = list(map(colors.get, y))\n",
    "\n",
    "plt.scatter(X[:, 0], X[:, 1], c=y_col, marker='o', s=20, edgecolors='k')\n",
    "plt.xlabel('Feature 1')\n",
    "plt.ylabel('Feature 2')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "6a75fba7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(420, 2) (180, 2)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split \n",
    "\n",
    "y = y.astype(int)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X, y, test_size=0.3, random_state=42)\n",
    "\n",
    "print(X_train.shape, X_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "e9d411c3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:5 out of the last 14 calls to <function Model.make_predict_function.<locals>.predict_function at 0x2f8ac85e0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:5 out of the last 14 calls to <function Model.make_predict_function.<locals>.predict_function at 0x2f8ac85e0> triggered tf.function retracing. Tracing is expensive and the excessive number of tracings could be due to (1) creating @tf.function repeatedly in a loop, (2) passing tensors with different shapes, (3) passing Python objects instead of tensors. For (1), please define your @tf.function outside of the loop. For (2), @tf.function has reduce_retracing=True option that can avoid unnecessary retracing. For (3), please refer to https://www.tensorflow.org/guide/function#controlling_retracing and https://www.tensorflow.org/api_docs/python/tf/function for  more details.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6/6 [==============================] - 0s 5ms/step\n",
      "softplus: 0.45\n",
      "6/6 [==============================] - 0s 4ms/step\n",
      "elu: 0.7111111111111111\n",
      "6/6 [==============================] - 0s 4ms/step\n",
      "relu: 0.29444444444444445\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score \n",
    "\n",
    "activation_functions = [\"softplus\", \"elu\", \"relu\"]\n",
    "\n",
    "def create_model(activation):\n",
    "    \"\"\"\n",
    "    MLP with one hidden layer and different choices of\n",
    "    activation functions\n",
    "    \"\"\"\n",
    "    \n",
    "    inputs_ = keras.Input(shape=(2,), name='input')\n",
    "    \n",
    "    hidden = keras.layers.Dense(2, activation=activation, name='hidden')(inputs_)\n",
    "    \n",
    "    outputs_ = keras.layers.Dense(3, activation=\"softmax\", name='output')(hidden)\n",
    "    \n",
    "    model = keras.Model(inputs=inputs_, outputs=outputs_)\n",
    "    \n",
    "    return model\n",
    "\n",
    "keras.utils.set_random_seed(seed=4)\n",
    "\n",
    "for item in activation_functions:\n",
    "    \n",
    "    keras.backend.clear_session()\n",
    "    \n",
    "    model = create_model(item)\n",
    "    \n",
    "    # Compile \n",
    "    model.compile(loss=\"sparse_categorical_crossentropy\", \n",
    "              optimizer=\"sgd\",\n",
    "              metrics=[\"accuracy\"]) \n",
    "    \n",
    "    # Train \n",
    "    model.fit(X_train, y_train, epochs=40, batch_size=32, verbose=0)\n",
    "    \n",
    "    # Evaluate \n",
    "    print(str(item) + \": \" + str(accuracy_score(y_test, model.predict(X_test).argmax(axis=1))))\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3cbb517a",
   "metadata": {},
   "source": [
    "### Third example\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* Neural network with different activation functions\n",
    "\n",
    "* In this example, we will create a model with three branches, each using a different activation function, and then concatenate the outputs of these branches before the final output layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "b9fff1e2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6/6 [==============================] - 0s 7ms/step\n",
      "combined: 0.8888888888888888\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras.layers import Input, Dense, Concatenate\n",
    "\n",
    "# Define input shape\n",
    "input_dim = 2\n",
    "input_shape = (input_dim,)\n",
    "\n",
    "keras.utils.set_random_seed(seed=4)\n",
    "\n",
    "# Define the input layer\n",
    "input_layer = Input(shape=input_shape)\n",
    "\n",
    "# Create two branches with different activation functions\n",
    "branch_1 = Dense(2, activation='elu')(input_layer)\n",
    "branch_2 = Dense(2, activation='softplus')(input_layer)\n",
    "branch_3 = Dense(2, activation='relu')(input_layer)\n",
    "\n",
    "\n",
    "# Concatenate the outputs of the two branches\n",
    "concatenated = Concatenate()([branch_1, branch_2, branch_3])\n",
    "\n",
    "# Output layer\n",
    "output_dim = 3\n",
    "output_layer = Dense(output_dim, activation='softmax')(concatenated)\n",
    "\n",
    "# Create the model\n",
    "model = keras.Model(inputs=input_layer, outputs=output_layer)\n",
    "\n",
    "# Compile the model\n",
    "model.compile(optimizer='sgd', loss='sparse_categorical_crossentropy', \n",
    "              metrics=['accuracy'])\n",
    "\n",
    "# Train\n",
    "model.fit(X_train, y_train, epochs=40, batch_size=32, verbose=0)\n",
    "\n",
    " \n",
    "# Evaluate \n",
    "print(\"combined\" + \": \" + str(accuracy_score(y_test, model.predict(X_test).argmax(axis=1))))\n",
    "    \n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "879f7778",
   "metadata": {},
   "source": [
    "### Fourth example: trainable activation functions\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "\n",
    "* Fixed activation functions are chosen before training \n",
    "    * Remain unchanged throughout the training process\n",
    "\n",
    "* These fixed functions might not be optimal for all layers or tasks\n",
    "    * Leading to suboptimal performance\n",
    "\n",
    "* Trainable activation functions are neural network components that adapt and learn their parameters during training\n",
    "\n",
    "    * Instead of using a fixed mathematical function like ReLU, trainable activation functions learn their function's shape from data\n",
    "    * Trainable activation functions are often implemented as additional neural network layers "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "7ad9b7fe",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 32ms/step - loss: 0.1719 - accuracy: 1.0000\n",
      "Test accuracy with Custom Activation: 100.00%\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.datasets import load_iris\n",
    "from keras.layers import Dense\n",
    "\n",
    "tf.keras.utils.set_random_seed(5)\n",
    "\n",
    "# Load the Iris dataset\n",
    "data = load_iris()\n",
    "X, y = data.data, data.target\n",
    "\n",
    "# Standardize the features\n",
    "scaler = StandardScaler()\n",
    "X = scaler.fit_transform(X)\n",
    "\n",
    "# Split the dataset into training and testing sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "# Create a custom activation function with a learnable parameter\n",
    "class CustomActivation(tf.keras.layers.Layer):\n",
    "    def __init__(self, units=1, **kwargs):\n",
    "        super(CustomActivation, self).__init__(**kwargs)\n",
    "        self.units = units\n",
    "\n",
    "    def build(self, input_shape):\n",
    "        self.alpha = self.add_weight(\n",
    "            shape=(self.units,),\n",
    "            initializer='ones',\n",
    "            trainable=True,\n",
    "            name='alpha',\n",
    "        )\n",
    "        super(CustomActivation, self).build(input_shape)\n",
    "\n",
    "    def call(self, x):\n",
    "        return self.alpha * tf.keras.activations.relu(x) # relu\n",
    "\n",
    "# Create a Sequential model\n",
    "model_with_custom_activation = keras.Sequential()\n",
    "\n",
    "# Add a hidden layer with the custom activation function\n",
    "model_with_custom_activation.add(Dense(32, activation=CustomActivation(units=1)))\n",
    "\n",
    "# Add an output layer (3 classes in Iris dataset)\n",
    "model_with_custom_activation.add(Dense(3, activation='softmax'))\n",
    "\n",
    "# Compile the model with SGD optimizer\n",
    "optimizer = keras.optimizers.SGD(learning_rate=0.01)\n",
    "model_with_custom_activation.compile(loss='sparse_categorical_crossentropy', \n",
    "                                     optimizer=optimizer, metrics=['accuracy'])\n",
    "\n",
    "# Train the model\n",
    "model_with_custom_activation.fit(X_train, y_train, epochs=100, batch_size=16, \n",
    "                                 validation_split=0.1, verbose=0)\n",
    "\n",
    "# Evaluate the model on the test set\n",
    "test_loss, test_accuracy = model_with_custom_activation.evaluate(X_test, y_test)\n",
    "print(f'Test accuracy with Custom Activation: {test_accuracy * 100:.2f}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "0849e615",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense_4 (Dense)             (None, 32)                161       \n",
      "                                                                 \n",
      " dense_5 (Dense)             (None, 3)                 99        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 260\n",
      "Trainable params: 260\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model_with_custom_activation.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "14ae7098",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tf.Variable 'dense_4/custom_activation/alpha:0' shape=(1,) dtype=float32, numpy=array([1.7303293], dtype=float32)>"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_with_custom_activation.weights[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "9f76f2f8",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.SGD` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.SGD`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.SGD`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 37ms/step - loss: 0.2499 - accuracy: 0.9333\n",
      "Test accuracy without Custom Activation: 93.33%\n"
     ]
    }
   ],
   "source": [
    "tf.keras.utils.set_random_seed(5)\n",
    "\n",
    "\n",
    "# Create a Sequential model without the custom activation function\n",
    "model_without_custom_activation = keras.Sequential()\n",
    "\n",
    "\n",
    "# Add a hidden layer\n",
    "model_without_custom_activation.add(Dense(32, activation='relu'))\n",
    "\n",
    "# Add an output layer (3 classes in Iris dataset)\n",
    "model_without_custom_activation.add(Dense(3, activation='softmax'))\n",
    "\n",
    "# Compile the model with SGD optimizer\n",
    "optimizer = keras.optimizers.SGD(learning_rate=0.01)\n",
    "model_without_custom_activation.compile(loss='sparse_categorical_crossentropy', \n",
    "                                        optimizer=optimizer, metrics=['accuracy'])\n",
    "\n",
    "# Train the model\n",
    "model_without_custom_activation.fit(X_train, y_train, epochs=100, batch_size=16, \n",
    "                                    validation_split=0.1, verbose=0)\n",
    "\n",
    "# Evaluate the model on the test set\n",
    "test_loss, test_accuracy = model_without_custom_activation.evaluate(X_test, y_test)\n",
    "print(f'Test accuracy without Custom Activation: {test_accuracy * 100:.2f}%')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c25131bf",
   "metadata": {},
   "source": [
    "### Last example \n",
    "\n",
    "https://scikit-learn.org/stable/modules/generated/sklearn.datasets.fetch_california_housing.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "2d60bd1f",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn.datasets import fetch_california_housing\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from tensorflow.keras.layers import Input, Dense, concatenate\n",
    "from tensorflow.keras.models import Model\n",
    "\n",
    "# Load California housing dataset\n",
    "housing = fetch_california_housing()\n",
    "X, y = housing.data, housing.target\n",
    "\n",
    "ind = y < 4\n",
    "X = X[ind]\n",
    "y = y[ind]\n",
    "\n",
    "# Split the data into training and testing sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "# Standardize the input features\n",
    "scaler = StandardScaler()\n",
    "X_train_scaled = scaler.fit_transform(X_train)\n",
    "X_test_scaled = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "c6815371",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([False,  True,  True, ...,  True,  True,  True]), (20640,))"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ind, ind.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "6e866412",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(15095, 8)"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_scaled.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "ff4750de",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_2\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                   Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      " deep_input (InputLayer)        [(None, 6)]          0           []                               \n",
      "                                                                                                  \n",
      " hidden1 (Dense)                (None, 64)           448         ['deep_input[0][0]']             \n",
      "                                                                                                  \n",
      " shallow_input (InputLayer)     [(None, 2)]          0           []                               \n",
      "                                                                                                  \n",
      " hidden2 (Dense)                (None, 32)           2080        ['hidden1[0][0]']                \n",
      "                                                                                                  \n",
      " shallow_output (Dense)         (None, 1)            3           ['shallow_input[0][0]']          \n",
      "                                                                                                  \n",
      " deep_output (Dense)            (None, 1)            33          ['hidden2[0][0]']                \n",
      "                                                                                                  \n",
      " combined_output (Concatenate)  (None, 2)            0           ['shallow_output[0][0]',         \n",
      "                                                                  'deep_output[0][0]']            \n",
      "                                                                                                  \n",
      " output (Dense)                 (None, 1)            3           ['combined_output[0][0]']        \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 2,567\n",
      "Trainable params: 2,567\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Define the shallow part (linear model)\n",
    "shallow_input = Input(shape=(2,), name='shallow_input')\n",
    "shallow_output = Dense(1, activation='linear', name='shallow_output')(shallow_input)\n",
    "\n",
    "# Define the deep part (deep neural network)\n",
    "deep_input = Input(shape=(6,), name='deep_input')\n",
    "deep_hidden1 = Dense(64, activation='relu', name='hidden1')(deep_input)\n",
    "deep_hidden2 = Dense(32, activation='relu', name='hidden2')(deep_hidden1)\n",
    "deep_output = Dense(1, activation='linear', name='deep_output')(deep_hidden2)\n",
    "\n",
    "# Combine the wide and deep parts\n",
    "combined_output = concatenate([shallow_output, deep_output], name='combined_output')\n",
    "\n",
    "# Final output layer for regression\n",
    "output = Dense(1, activation='linear', name='output')(combined_output)\n",
    "\n",
    "# Create the model\n",
    "model = Model(inputs=[shallow_input, deep_input], outputs=output)\n",
    "\n",
    "# Compile the model\n",
    "model.compile(optimizer='adam', loss='mean_squared_error', metrics=['mae'])\n",
    "\n",
    "# Display the model summary\n",
    "model.summary()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "13d35cb4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from tensorflow.keras.utils import plot_model\n",
    "\n",
    "plot_model(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "e15404ac",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "236/236 [==============================] - 2s 6ms/step - loss: 1.8268 - mae: 1.0297\n",
      "Epoch 2/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.6113 - mae: 0.6014\n",
      "Epoch 3/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.4323 - mae: 0.5043\n",
      "Epoch 4/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.3541 - mae: 0.4500\n",
      "Epoch 5/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.3252 - mae: 0.4183\n",
      "Epoch 6/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2933 - mae: 0.3966\n",
      "Epoch 7/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2653 - mae: 0.3825\n",
      "Epoch 8/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2577 - mae: 0.3748\n",
      "Epoch 9/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2512 - mae: 0.3681\n",
      "Epoch 10/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2503 - mae: 0.3652\n",
      "Epoch 11/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2526 - mae: 0.3622\n",
      "Epoch 12/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2394 - mae: 0.3570\n",
      "Epoch 13/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2328 - mae: 0.3534\n",
      "Epoch 14/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2277 - mae: 0.3498\n",
      "Epoch 15/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2229 - mae: 0.3467\n",
      "Epoch 16/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2218 - mae: 0.3432\n",
      "Epoch 17/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2294 - mae: 0.3427\n",
      "Epoch 18/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2172 - mae: 0.3366\n",
      "Epoch 19/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2095 - mae: 0.3325\n",
      "Epoch 20/20\n",
      "236/236 [==============================] - 1s 5ms/step - loss: 0.2031 - mae: 0.3275\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x2f8ae4340>"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Train the model\n",
    "model.fit([X_train_scaled[:,:2], X_train_scaled[:,2:]], y_train, epochs=20, batch_size=64)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "2a7d0a9c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "118/118 [==============================] - 0s 2ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Make predictions \n",
    "from sklearn.metrics import r2_score\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "y_pred = model.predict([X_test_scaled[:,:2], X_test_scaled[:,2:]])\n",
    "plt.plot(y_test, y_pred, 'o')\n",
    "plt.plot([0,5], [0, 5], 'r--')\n",
    "plt.xlabel('True values')\n",
    "plt.ylabel('Predicted values')\n",
    "plt.title(f'r2 score is {r2_score(y_test, y_pred).round(3)}')\n",
    "plt.xlim([0,5.2])\n",
    "plt.ylim([0,5.2])\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "185a5ba1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.574"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# What if we used linear regression? \n",
    "\n",
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "lm = LinearRegression()\n",
    "\n",
    "lm.fit(X_train_scaled, y_train)\n",
    "\n",
    "y_pred = lm.predict(X_test_scaled)\n",
    "\n",
    "r2_score(y_test, y_pred).round(3)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "059a54e6",
   "metadata": {},
   "source": [
    "### Convolutional neural networks (CNNs)\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* CNN captures the spatial features from an image, i.e., the arrangement of pixels and the relationship between them\n",
    "\n",
    "* They have three main types of layers, which are:\n",
    "    * Convolutional layer\n",
    "    * Pooling layer\n",
    "    * Fully-connected (FC) or Dense layer \n",
    "    \n",
    "* Advantages over MLPs\n",
    "    * Fewer parameters to learn\n",
    "    * All patches of an image will be treated in the same manner (translation invariance)\n",
    "    * Only a small neighborhood of pixels will be used to compute hidden representations (locality)\n",
    "   "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c45faa97",
   "metadata": {},
   "source": [
    "* A convolution layer is made up of a large number of convolution filters \n",
    "\n",
    "* A convolution filter relies on a simple operation, called a convolution\n",
    "\n",
    "    * repeatedly multiplying matrix elements and then adding the results\n",
    "    * consider a very simple example of a 4x3 image\n",
    "$$\n",
    "\\text{image}=\\begin{bmatrix}\n",
    "    a_{11} & a_{12} & a_{13} \\\\\n",
    "    a_{21} & a_{22} & a_{23} \\\\\n",
    "    a_{31} & a_{32} & a_{33} \\\\\n",
    "    a_{41} & a_{42} & a_{43} \\\\\n",
    "\\end{bmatrix}\n",
    "$$\n",
    "* consider a 2x2 filter of the form\n",
    "$$\n",
    "\\text{filter}=\\begin{bmatrix}\n",
    "    f_{11} & f_{12} \\\\\n",
    "    f_{21} & f_{22} \\\\\n",
    "\\end{bmatrix}\n",
    "$$\n",
    "\n",
    "* When we convolve the image with the filter, we get the result\n",
    "$$\n",
    "\\begin{bmatrix}\n",
    "  a_{11}f_{11}+a_{12}f_{12}+a_{21}f_{21}+a_{22}f_{22}   &  a_{12}f_{11}+a_{13}f_{12}+a_{22}f_{21}+a_{23}f_{22}\\\\\n",
    "   a_{21}f_{11}+a_{22}f_{12}+a_{31}f_{21}+a_{32}f_{22}  &  a_{22}f_{11}+a_{23}f_{12}+a_{32}f_{21}+a_{33}f_{22}\\\\\n",
    "    a_{31}f_{11}+a_{32}f_{12}+a_{41}f_{21}+a_{42}f_{22} &  a_{32}f_{11}+a_{33}f_{12}+a_{42}f_{21}+a_{43}f_{22}\\\\\n",
    "\\end{bmatrix}\n",
    "$$\n",
    "\n",
    "* If the image size is $n_h\\times n_w$ and the size of the convolution filter is $k_h\\times k_w$, we get \n",
    "\n",
    "$$\\big(n_h - k_h + 1 \\big)\\times \\big(n_w - k_w + 1 \\big)$$\n",
    "\n",
    "* Thus, the output size is slightly smaller than the input size\n",
    "\n",
    "* \"Padding\" refers to the technique of adding extra rows and columns of zeros around the input data before applying a convolution operation\n",
    "    * control the spatial dimensions of the output feature maps\n",
    "    * preserve information at the edges of the input\n",
    "    \n",
    "* If we add a total of $p_h$ rows of padding and $p_w$ columns of padding, the output shape will be\n",
    "\n",
    "$$\\big(n_h - k_h + p_h + 1 \\big)\\times \\big(n_w - k_w + p_w + 1 \\big)$$\n",
    "\n",
    "* Hence, we can give the input and output the same height and width by choosing \n",
    "\n",
    "$$p_h = k_h - 1, \\; p_w = k_w - 1$$\n",
    "\n",
    "* This is why CNNs commonly use convolution filters with odd height and width values\n",
    "    * padding with the same number of rows on top and bottom, and the same number of columns on left and right\n",
    "    \n",
    "* \"Stride\" controls how much the filter shifts (horizontally and vertically) between successive applications to generate the output feature map\n",
    "    * That is, the number of rows and columns traversed per slide \n",
    "    * A stride of 1 means that the filter moves one pixel at a time (no skipping)\n",
    "    * A stride greater than 1 causes the filter to skip pixels\n",
    "    \n",
    "$$\\lfloor\\big(n_h-k_h+p_h+s_h\\big)/s_h\\rfloor \\times \\lfloor\\big(n_w-k_w+p_w+s_w\\big)/s_w\\rfloor$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a4494652",
   "metadata": {},
   "source": [
    "### Multiple input channels\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* When the input data or image contains multiple channels, we need to construct a convolution filter with the same number of channels \n",
    "\n",
    "    * For example, consider $C$ channels for the previous case study, $i=1,2,\\ldots,C$\n",
    " \n",
    "    $$\\text{channel $i$ of image} =\\begin{bmatrix}\n",
    "    a_{11}^{(i)} & a_{12}^{(i)} & a_{13}^{(i)} \\\\\n",
    "    a_{21}^{(i)} & a_{22}^{(i)} & a_{23}^{(i)} \\\\\n",
    "    a_{31}^{(i)} & a_{32}^{(i)} & a_{33}^{(i)} \\\\\n",
    "    a_{41}^{(i)} & a_{42}^{(i)} & a_{43}^{(i)} \\\\\n",
    "\\end{bmatrix}$$\n",
    "\n",
    "$$\\text{channel $i$ of filter}=\\begin{bmatrix}\n",
    "    f_{11}^{(i)} & f_{12}^{(i)} \\\\\n",
    "    f_{21}^{(i)} & f_{22}^{(i)} \\\\\n",
    "\\end{bmatrix}$$\n",
    "\n",
    "* Final result: two-dimensional tensor\n",
    "\n",
    "$$\\begin{bmatrix}\n",
    "  \\sum_{i=1}^C [a_{11}^{(i)}f_{11}^{(i)}+a_{12}^{(i)}f_{12}^{(i)}+a_{21}^{(i)}f_{21}^{(i)}+a_{22}^{(i)}f_{22}^{(i)}]   &  \\sum_{i=1}^C [a_{12}^{(i)}f_{11}^{(i)}+a_{13}^{(i)}f_{12}^{(i)}+a_{22}^{(i)}f_{21}^{(i)}+a_{23}^{(i)}f_{22}^{(i)}]\\\\\n",
    "\\sum_{i=1}^C   [a_{21}^{(i)}f_{11}^{(i)}+a_{22}^{(i)}f_{12}^{(i)}+a_{31}^{(i)}f_{21}^{(i)}+a_{32}^{(i)}f_{22}^{(i)}]  & \\sum_{i=1}^C [a_{22}^{(i)}f_{11}^{(i)}+a_{23}^{(i)}f_{12}^{(i)}+a_{32}^{(i)}f_{21}^{(i)}+a_{33}^{(i)}f_{22}^{(i)}]\\\\\n",
    "    \\sum_{i=1}^C [a_{31}^{(i)}f_{11}^{(i)}+a_{32}^{(i)}f_{12}^{(i)}+a_{41}^{(i)}f_{21}^{(i)}+a_{42}^{(i)}f_{22}^{(i)}] &  \\sum_{i=1}^C [a_{32}^{(i)}f_{11}^{(i)}+a_{33}^{(i)}f_{12}^{(i)}+a_{42}^{(i)}f_{21}^{(i)}+a_{43}^{(i)}f_{22}^{(i)}]\\\\\n",
    "\\end{bmatrix}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f489bbc6",
   "metadata": {},
   "source": [
    "### Multiple output channels\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* In a convolution layer, you typically have multiple filters. Each filter is a small, learnable tensor that slides or convolves over the input data\n",
    "\n",
    "* The number of filters you specify in a layer determines the number of output channels\n",
    "\n",
    "* Denote by $C_i$ and $C_o$ the number of input and output channels\n",
    "\n",
    "    * Let $k_h$ and $k_w$ be the height and width of the filter\n",
    "    * To get an output with multiple channels, we can create a filter tensor of shape $k_h\\times k_w \\times C_i$ for every output channel \n",
    "    * Hence, we concatenate them on the output channel dimension, so that the shape of the convolution filter is $k_h\\times k_w \\times C_i \\times C_o$\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2c8e6096",
   "metadata": {},
   "source": [
    "### Maximum pooling\n",
    "<hr style=\"border:2px solid gray\">\n",
    "\n",
    "* The primary purpose of max pooling is to downsample the spatial dimensions of feature maps\n",
    "\n",
    "* This reduction in spatial resolution helps control the computational complexity of the network \n",
    "\n",
    "* Max pooling introduces a degree of position invariance, meaning that the model can recognize features regardless of their precise location in the input\n",
    "\n",
    "* In each pooling window (typically 2x2 or 3x3), max pooling selects the maximum value\n",
    "\n",
    "    * It operates on each channel (feature map) of the input independently\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9c1648b3",
   "metadata": {},
   "source": [
    "Example: https://github.com/yu4u/convnet-drawer\n",
    "\n",
    "Conv2D: https://keras.io/api/layers/convolution_layers/convolution2d/\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cba795ab",
   "metadata": {},
   "source": [
    "### Example\n",
    "\n",
    "* Data set: https://github.com/rmccorm4/Tiny-Imagenet-200 \n",
    "* Consider 3 classes only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "ae6fa196",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1500 1500\n",
      "./images/n01629819/n01629819_91.JPEG 1\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "from PIL import Image\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "# Set the path to the main directory containing the 3 folders\n",
    "main_directory = './images'\n",
    "\n",
    "# Initialize lists to store image paths and corresponding labels\n",
    "image_paths = []\n",
    "labels = []\n",
    "\n",
    "# Iterate through each folder\n",
    "for folder_name in os.listdir(main_directory):\n",
    "    folder_path = os.path.join(main_directory, folder_name)\n",
    "\n",
    "    # Ensure that it's a directory\n",
    "    if os.path.isdir(folder_path):\n",
    "        # Assign a label based on the folder name\n",
    "        label = folder_name\n",
    "\n",
    "        # Iterate through each image in the folder\n",
    "        for filename in os.listdir(folder_path):\n",
    "            if filename.endswith(('.JPEG')):  # Adjust file extensions as needed\n",
    "                image_path = os.path.join(folder_path, filename)\n",
    "\n",
    "                # Append the image path and label to the lists\n",
    "                image_paths.append(image_path)\n",
    "                labels.append(label)\n",
    "\n",
    "# Convert labels to numerical format using LabelEncoder\n",
    "label_encoder = LabelEncoder()\n",
    "encoded_labels = label_encoder.fit_transform(labels)\n",
    "\n",
    "print(len(image_paths), len(encoded_labels))\n",
    "\n",
    "print(image_paths[10], encoded_labels[10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "75bbb3a6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'RGB'"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "image = Image.open(image_paths[10])\n",
    "\n",
    "image.show()\n",
    "\n",
    "image.mode"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "da86f2b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split the image into its RGB channels\n",
    "r, g, b = image.split()\n",
    "\n",
    "# Convert each channel to NumPy array\n",
    "r_array = np.array(r)\n",
    "g_array = np.array(g)\n",
    "b_array = np.array(b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "f405c415",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[238, 238, 238, ..., 233, 231, 245],\n",
       "       [238, 238, 238, ..., 238, 229, 240],\n",
       "       [238, 238, 238, ..., 246, 225, 226],\n",
       "       ...,\n",
       "       [237, 237, 237, ..., 238, 238, 238],\n",
       "       [237, 237, 237, ..., 238, 238, 238],\n",
       "       [237, 237, 237, ..., 238, 238, 238]], dtype=uint8)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r_array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "9d2ec956",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1495 1495\n"
     ]
    }
   ],
   "source": [
    "# We have some non RGB images \n",
    "\n",
    "image_paths_rgb = []\n",
    "indices_rgb = []\n",
    "\n",
    "for i in range(1500):\n",
    "    image = Image.open(image_paths[i])\n",
    "    if image.mode == 'RGB':\n",
    "        image_paths_rgb.append(image_paths[i])\n",
    "        indices_rgb.append(i)\n",
    "        \n",
    "print(len(image_paths_rgb), len(indices_rgb))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "8f42f3b9",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = np.zeros((1495, 64, 64, 3))\n",
    "\n",
    "for i in range(1495):\n",
    "    image = Image.open(image_paths_rgb[i])\n",
    "    r, g, b = image.split()\n",
    "    X[i, :, :, 0] = np.array(r)/255.\n",
    "    X[i, :, :, 1] = np.array(g)/255.\n",
    "    X[i, :, :, 2] = np.array(b)/255.\n",
    "\n",
    "\n",
    "# Split the data into training and testing sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, encoded_labels[indices_rgb], \n",
    "                                                    test_size=0.25, random_state=1)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "19d8f735",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf \n",
    "\n",
    "from tensorflow import keras \n",
    "\n",
    "from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense\n",
    "\n",
    "from keras.models import Sequential \n",
    "\n",
    "model = Sequential()\n",
    "\n",
    "model.add(Conv2D(20, (3, 3), (2, 2), activation='relu', input_shape=(64, 64, 3)))\n",
    "model.add(MaxPooling2D((3, 3), strides=(2, 2)))\n",
    "model.add(Conv2D(50, (5, 5), activation='relu', padding=\"same\"))\n",
    "model.add(MaxPooling2D((3, 3), strides=(2, 2)))\n",
    "model.add(Flatten())\n",
    "model.add(Dense(32, activation='relu'))\n",
    "model.add(Dense(16, activation='relu'))\n",
    "model.add(Dense(3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "2f2879c9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_2\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 31, 31, 20)        560       \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 15, 15, 20)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 15, 15, 50)        25050     \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 7, 7, 50)         0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 2450)              0         \n",
      "                                                                 \n",
      " dense_8 (Dense)             (None, 32)                78432     \n",
      "                                                                 \n",
      " dense_9 (Dense)             (None, 16)                528       \n",
      "                                                                 \n",
      " dense_10 (Dense)            (None, 3)                 51        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 104,621\n",
      "Trainable params: 104,621\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "7f74acfb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAASkAAANHCAYAAABq+KuBAAAABmJLR0QA/wD/AP+gvaeTAAAgAElEQVR4nOzdeVQUZ9o28KtoVhvBFRAxQkRUXCJJXFCIeIgE0Diyicomvm9M3CI6ijoTj+iYMcZJnBySiQlRwAyvIkYwwbjHMEE0GuOSBBdAVHADEVCRpWnu7w8/amxpsGlaumjv3zmcYz/9VNdd5dMX9VQV3QIRERhjTKKM9F0AY4y1hEOKMSZpHFKMMUnjkGKMSZqxvgtQZ/HixSgqKtJ3GYw9dzZu3AgHBwd9l6FCkkdS+/btQ25urr7LYM9Ibm4uDhw4oO8y2GPu3buHnTt3orKyUt+lNCHJIykACAkJQVxcnL7LYM9AXFwcduzYgbS0NH2Xwv6/P/74A0OGDNF3GWpJ8kiKMcYacUgxxiSNQ4oxJmkcUowxSeOQYoxJGocUY0zSOKQYY5LGIcUYkzQOKcaYpHFIMcYkjUOKMSZpHFKMMUnjkGKMSRqHFGNM0jikGGOSxiHFGJM0Dikdy83NhaWlJfr06QNTU1MIgoBXXnkFDx48aHaZI0eOYOzYsTA2bt1nEN6/fx+enp745ptv2lq21mpra5/5Oi5cuIAuXbpAEAR8/fXXUCqVz3yd6jy5rT/99JP4f/ztt9/i4cOHeqnL0HFI6djmzZtx6NAhFBUVobCwEKGhofj111/x/vvvN7vM+PHjMXbs2Favq3Pnzvjpp58QFBTUlpLb5K9//SsaGhqe6ToGDhwIZ2dnCIKAsLAwyGSyZ7q+5jy5rZ6enujduze6d++OyZMno1OnTnqpy9BxSOlQRUUFPDw8MHr0aABA7969sX79egiCgJ9//rnFZU1MTNqjRJ367bffsGnTpnZZl7m5OWQyGYyM9DNkm9tWU1NTmJqa6qGi54dkP+O8tWpqapCWloY//vgD7u7u8PHxgYWFhfh8bW0tsrKykJWVBXt7e/j6+qJfv37i8/n5+UhKSsKaNWtQUFCAHTt2wMbGBjNnzoSJiQn27NmDe/fuAQAEQUBAQADMzMzw66+/4uLFiwCAiRMnIiAgQKWuvn37YvDgwejfv79Ku0KhQHp6Ok6fPg0vLy+tjkYat9nW1hY+Pj4abQcAFBQU4LvvvkNMTAyys7Oxd+9euLi4ICIiAkZGRkhNTUVDQwNMTEwQHBwMANi5cycUCgUsLCwwZcoUHD16FDNmzEBVVRW2b98OExMThISEtHob2qKjbWteXh6+//57VFRUYOTIkfDz8wMA7N69W5wqCoKAadOmAXj0uePnzp0DAPj4+KB79+64d+8eUlNTcf78ebz44ouYOXMmLC0txW1NSkpCXFwc9u7di9zcXCxatKhD/gJUQRI0aNAgWrVqlcb9r1y5Qp6enpSQkEBFRUXk7e1NL774IlVXVxMRUXV1NXl5edH27dupvLyc4uPjqXPnzvTNN98QEVFSUhLZ2toSAPr2228pMDCQJk6cSABo5cqVRER08+ZNeuWVVwgAZWdni+tuaGggf39/2rZtm9ralEolyeVycV1ERBUVFeTt7U1xcXFUVlZGycnJZGpqSjKZTONtPn/+PE2ZMoUA0Pr16zXejvj4eLK0tKRevXpRSkoKDR06lCwsLAgABQUFERHRvXv3aOzYsWRlZSWu78aNGzR06FCys7MjIqKffvqJwsLCCABlZmbS/v37Na591apVNGjQII37ExGNHTuWjI2NxcdS2VYXFxfq1avXU+tfsGABeXp60p07d+jAgQMkCAJ98MEHRPTo/7JXr14EgPLy8sRllEoleXt706effkoNDQ106dIlevPNN2n//v105swZGjJkCPXr14/Ky8spOTmZ7OzsCAAlJSWRm5sbAaCjR49qtH9///13AkC///67Rv3bk0GE1IQJE2j27Nni48zMTBIEgXbt2kVERDNmzKDo6GiVZYKDg8nCwoKKioqIiCg2NpYA0O7du8U+48ePJxcXF/HxwYMHCQClpKSIbbW1tRQYGNhsbenp6TRq1ChqaGgQ2+bOnUtTpkxR6Tdp0qRWhRQR0fXr11VCStPtCA0NJblcTv/+97+J6NGb0t3dnQCIb8D58+ervHGJiP73f/9XfOMSEa1evZoAqGybJnQRUkTS2FZNQ8ra2prWrl0rPnZ1daXRo0eLj1NSUlRqIiKqq6ujV199lerr64no0ThPT08Xn9+7d69KKP/1r38VQ4qI6MKFCxr/30g5pAzinNTBgwcxZcoU8fHEiRNx69YtBAQE4OHDh0hLS4Obm5vKMnPmzEF1dTUSExMBAHK5HADg7+8v9hkyZAiKi4vFx6+//joGDRqkcm5i165dzR76KxQKrFu3DsnJyRAEAQBQUlKChIQEcXrWaNiwYa3e7sbD/Mdpsh1yuRxWVlYICwsDAPTq1Qvr1q0D8GhfAlB77kdf54Oa05G2dc+ePZgzZw4A4MSJEyAiVFdXi8+HhobC2dkZ//jHP8S29PR0TJkyBTKZDDdv3sTBgweRk5ODFStWYMWKFdizZw9effVVcarYeHpj+vTpAIABAwaI464jM5hzUk++YW1sbAAAOTk5UCgUTS7vN54junTpEgD1g1Iul6O+vl6lbf78+Zg3bx7OnTuHYcOGIT09HV9//bXammJiYrBq1SoMGDBAbDt79iwUCgXs7OxU+mozmDR9c6nbjifXN2LECADoUN8c3ZG2dezYsUhPT8euXbvwxhtvwNHREdevXxefl8lkWLZsGd566y2cOHECI0eOxObNm5GcnAzg0fksAIiNjUWPHj3UrsMQAkkdaf1qbIP9+/c3aSstLRXvqcnJyVF5rvE/2sXFpVXriYyMhJWVFT799FOcP38e/fv3V3t155NPPsGIESNUfssDj+5tAoCbN2+2ar3PmqmpKczMzPDCCy/ou5Rnrj239fFw2bJlCxISEhAeHg4zM7MmfSMjI9G7d2+8//77uHjxIrp06SL+MmscY7/++muT5RrHlKEyiJAyMjLCd999p3KTX0FBAX755Re4ubnBzMwMR48eVVmmtLQUwKN7XVrD0tISUVFRSElJwYYNGzB79uwmfbZs2QJBEDBz5kyxjYhw4cIFDBw4EACwd+/eJss96/uNHldTU6PyOCcnB7W1tRg5ciQAwMrKqsnNi0Sk9kZKfd1cqalnva1EpHa9DQ0NSEhIwKlTp7BhwwbMmzcP5ubmzS5namqKJUuWiFcj33nnHfG5AQMGQCaTYdWqVairqxPbS0tLkZKS0tymGwSDCKnIyEicO3cOISEh+OGHH/DZZ59h5cqV8PX1hY2NDRYsWIDCwkIcOXJEXCYjIwMhISEYN24cAODu3bsAoHKeoL6+HgqFoskAnjdvHqqrq1FWVtbkt/GmTZvw1VdfwcrKCklJSUhMTER8fDwmTZqE0tJSuLq6wtfXF5mZmUhKSgIA1NXV4cyZMyAiFBUVNZmuNKfxLvaqqiqxTdPtqKysxLVr18TH+/btw6uvvireGNq3b1/U1tbi4MGDICKkpqYiJycHlZWVqKyshFKpRM+ePQEAp06dwk8//dQkDHTp/v37qK+vV7lzXwrbevPmTdy5c6fJGKmtrcW7774LR0dH8SbPjIwM1NfX49ChQzh79izKy8uRl5eHwsJCcbm33noL3bt3R2FhIcaPHy+2d+3aFe+88w6OHz+OcePG4f/+7/+QlJSEsLAw8RyUQqEAAJSVlWm7m6VJb6fsW9Daq3uVlZUUEBBAAAgAOTo60smTJ8XnlUolLV68mHr27EnLli2jqKgomjp1qniLQkZGBjk6OhIAWrhwIV2+fJm2b99OTk5OBICWLl1Kt2/fVlmnj48P7du3T6UtMTFRrOHJHycnJ/FKy61bt8jT05MAkIuLC02ePJnCw8PJ0tKS5s+fT8XFxU/d5mvXrtGcOXMIALm6utLevXs13o5Zs2aRXC6nyZMn02effUazZ88mDw8PKiwsFF+/qqqKhgwZQgDI1taWkpOTafbs2dS1a1dasmQJ3blzhy5fvky2trbUtWtX+uqrrzT+/2rt1b358+eTkZERAaCwsDA6cOCA3rf1+PHj4m0JAMjBwYFGjBhBI0eOpGHDhlHnzp1JEATx/zIiIoKMjIzI1taWNm3aRGvXriUjIyNasmRJk+2NjY2ljz/+uEl7VVUVRUZGiuu0srISr/bt3LmTBgwYQAAoJCSEzp49q/H+JZL21T2DCKlG169fpzNnzlBdXZ3a5x8+fEi//vqrGE5tce3atVZfen9Sfn4+Xbx4kRoaGujy5ctUWVnZ5ro0MWvWLLK3t6fa2lo6ffo0Xb58WW2/hoYGOnfuHFVVVRER0aVLl+jhw4cqferq6pq0PY02tyBoS9/b+riSkhKVsXn37l21/fz9/Zt9joiotLSUTp061aZaniTlkDKYq3sAYG9vD3t7+2aft7CwaHIrgrb69OnT5td4/I53JyenNr9ea5mammL48OHNPi8IAoYOHSo+fvKueeDRn/N0hDuapbCtjVPGRl27dm3SJycnB3369FH7XKMePXo0e4XPEBlUSDHNPHz4UOU8liHrCNt64sQJLF68GIMHD0Zubi4yMzP1XZKkcEhJTFFREaKjo5/aLyoqChEREa16bYVCgYSEBGRlZeH+/ftYuXIl3n77bTg4OGhbrmR1tG3Ny8uDhYUF/vnPf8La2lrf5UgKh5TEODg4YM+ePU/t19rPngIeTVfmzp2LuXPnalNah9KRtnXkyJG4ffu2vsuQLA4piREEQe2Nfow9rwziPinGmOHikGKMSRqHFGNM0jikGGOSxiHFGJM0DinGmKRxSDHGJI1DijEmaRxSjDFJ45BijEkahxRjTNIk+7d7jd9GzDqW8vJyyOXyFr96PDc3F9evX2/3bzxmzWv8dm4pkuSRlK+vL1xdXfVdBtPC4cOHUVJS0mIfV1fXJt87yPTLysoKwcHBkvyYGIGoma+6YEwLgiAgNTUVU6dO1XcpzEBI8kiKMcYacUgxxiSNQ4oxJmkcUowxSeOQYoxJGocUY0zSOKQYY5LGIcUYkzQOKcaYpHFIMcYkjUOKMSZpHFKMMUnjkGKMSRqHFGNM0jikGGOSxiHFGJM0DinGmKRxSDHGJI1DijEmaRxSjDFJ45BijEkahxRjTNI4pBhjksYhxRiTNA4pxpikcUgxxiSNQ4oxJmkcUowxSeOQYoxJGocUY0zSOKQYY5LGIcUYkzRjfRfAOq7Vq1fjxo0bTdoTEhJw+PBhlba4uDj06tWrvUpjBkQgItJ3EaxjWrJkCT766COYmJiIbQ0NDRAEAYIgAACUSiXs7OxQXFwstjHWGjzdY1qbPn06AEChUIg/SqUS9fX14mOZTIaoqCgOKKY1PpJibeLs7IyCgoIW+5w7dw5Dhw5tp4qYoeEjKdYm4eHhKtO9Jzk7O3NAsTbhkGJtEh4eDoVCofY5ExMTREdHt3NFzNDwdI+12UsvvYTffvsN6oZSXl4enJ2d9VAVMxR8JMXaLDIyEjKZTKVNEAS88sorHFCszTikWJtNnz4dDQ0NKm0ymQyRkZF6qogZEg4p1mb29vYYM2YMjIz+O5yUSiWCg4P1WBUzFBxSTCciIiLEf8tkMnh5ecHe3l6PFTFDwSHFdCIkJETlSOrx0GKsLTikmE507doVPj4+EAQBRkZGCAgI0HdJzEBwSDGdCQ8PBxHBz88PXbp00Xc5zEDwpyCoUVxcjJycHH2X0eHU19fDzMwML774Inbs2KHvcjqcPn36wN3dXd9lSA7fzKnGjh07EBoaqu8y2HMmODgYaWlp+i5DcvhIqgWc308nCAJSU1MxdepUAEBlZSWsra31XFXHExISou8SJIvPSTGd4oBiusYhxRiTNA4pxpikcUgxxiSNQ4oxJmkcUowxSeOQYoxJGocUY0zSOKQYY5LGIcUYkzQOKcaYpHFIMcYkjUOKMSZpHFKMMUnjkGKMSRqHFGNM0vhD7wzU1atXkZKSgpKSEgwfPhxhYWEwMTFpsf+XX36JlJQUXLlypV1qvH79OtLS0pCXl4fu3bvDy8sL7u7uuHbtGpRKJVxdXduljse1tN+OHj2KwsJClf7GxsawtrZGt27dMHToUHTq1KndazZ0HFIGKDc3FyNHjkTXrl1x+/ZtKBQKxMfHIysrC5aWlmqXuXz5Mn788UcUFxe3S43r1q3Dv/71L8yfPx+zZ8+Gg4MDfvnlF/j5+SE3Nxdffvllu4fU0/bbmDFj8ODBA/j5+cHa2hrvvvsunJyccPnyZZw8eRL79u3D+PHj8fHHH2PgwIHtWrtBI9ZEamoqdeRds3jxYjp27BgRERUXF1NoaCgBoOXLl7e43NKlS0kmk7VqXQAoNTW1Vcts3LiRTExMKDs7u8lz1dXVNGrUKNqyZUurXlMXNN1v3bp1owEDBjRZ/tChQ2RnZ0fm5uZ0/PjxVq07ODiYgoODtS/egPE5KQNTUVEBDw8PjB49GgDQu3dvrF+/HoIg4Oeff25x2Zamg7qybds2LFq0CH/7298wduzYJs+bm5sjPj4ed+/efea1PK41+83U1FTta3h7e2Pz5s2oqalBUFAQamtrn3ndzwOe7ulQTU0N0tLS8Mcff8Dd3R0+Pj6wsLAQn6+trUVWVhaysrJgb28PX19f9OvXT3w+Pz8fSUlJWLNmDQoKCrBjxw7Y2Nhg5syZMDExwZ49e3Dv3j0Aj74AISAgAGZmZvj1119x8eJFAMDEiRObfDFn3759MXjwYPTv31+lXaFQID09HadPn4aXlxcaGhqe1a4RrV27FgDg7+/fbJ8RI0aofAmG1PZbS/z9/eHt7Y3Dhw8jLS0N4eHhGi/LmqHvQzkp0ma6d+XKFfL09KSEhAQqKioib29vevHFF6m6upqIHk1jvLy8aPv27VReXk7x8fHUuXNn+uabb4iIKCkpiWxtbQkAffvttxQYGEgTJ04kALRy5UoiIrp58ya98sorBEBlqtTQ0ED+/v60bds2tbUplUqSy+XiuoiIKioqyNvbm+Li4qisrIySk5PJ1NT0mU738vLyCAAZGxtTbW2tRstIbb8REdnZ2amd7jV67733CADNmjVLo20k4uleSzik1NAmpCZMmECzZ88WH2dmZpIgCLRr1y4iIpoxYwZFR0erLBMcHEwWFhZUVFRERESxsbEEgHbv3i32GT9+PLm4uIiPDx48SAAoJSVFbKutraXAwMBma0tPT6dRo0ZRQ0OD2DZ37lyaMmWKSr9JkyY905BqrH3gwIEav77U9hvR00Nq69atBIAmTJig2UYSh1RL+JyUDuTn5+PgwYOYMmWK2DZx4kTcunULAQEBePjwIdLS0uDm5qay3Jw5c1BdXY3ExEQAgFwuB6A6FRoyZIjKFbfXX38dgwYNwqZNm8S2Xbt2Nfu9bQqFAuvWrUNycjIEQQAAlJSUICEhAT4+Pip9hw0bps3ma8zYuHVnF6S23zRVVVUFAOjZs2erlmPqcUjpwPnz5wGgyeV9GxsbAEBOTg4UCkWTN2njuY5Lly4BAIyMmv53yOVy1NfXq7TNnz8fP/30E86dOwcASE9PR2BgoNraYmJisGrVKgwYMEBsO3v2LBQKBezs7FT6tvbN2FqN25uXl4fq6uqn9pfaftNU43kufdznZYg4pHSg8eT4/v37mzxXWloKpVIJ4NGb7nE9evQAALi4uLRqfZGRkbCyssKnn36K8+fPo3///mqvOH3yyScYMWJEk5PU9+/fBwDcvHmzVettq969e2Pw4MFQKpX4/fffW+xLRJLbb5qoq6tDZmYmjI2Nm5yIZ9rhkNIBV1dXGBkZ4bvvvhPfWABQUFCAX375BW5ubjAzM8PRo0dVlistLQUAeHp6tmp9lpaWiIqKQkpKCjZs2IDZs2c36bNlyxYIgoCZM2eKbUSECxcuiDca7t27t8lyz/oK3+rVqwEAsbGxqKurU9untLQUiYmJkttvmtiwYQPy8/MRExPDR1I6wiGlA/b29oiMjMS5c+cQEhKCH374AZ999hlWrlwJX19f2NjYYMGCBSgsLMSRI0fE5TIyMhASEoJx48YBgHhv0ONTofr6eigUiib33MybNw/V1dUoKyvDCy+8oPLcpk2b8NVXX8HKygpJSUlITExEfHw8Jk2ahNLSUri6usLX1xeZmZlISkoC8OgI4MyZMyAiFBUVNZkq6UpQUBDWrl2LrKwsvPXWW+L5m0bXrl3DBx98gIiICMntN+DRuarGfz+utrYWixYtwurVq7FixQrxVgumA/o9by9N2lzdq6yspICAAAJAAMjR0ZFOnjwpPq9UKmnx4sXUs2dPWrZsGUVFRdHUqVPFWxQyMjLI0dGRANDChQvp8uXLtH37dnJyciIAtHTpUrp9+7bKOn18fGjfvn0qbYmJiWINT/44OTmJV6pu3bpFnp6eBIBcXFxo8uTJFB4eTpaWljR//nwqLi7WaLuhxR3nREQ7d+6kIUOGkIWFBY0bN47mzZtHb731FsXFxancniCl/faf//yHgoKCxNso3NzcKCAggIKCgmjSpEn0zjvv0KlTp1q9L4j46l5LBKLH7ppjAIAdO3YgNDQU2uyaGzduiEcr6u7grq6uxoULFzBo0CCYm5u3qc6ioiI4ODi06YR3QUEBlEol+vfvjytXrqB79+6wsrLSeHlBEJCamoqpU6dqtf6qqir89ttvcHZ2Fs81qSO1/aZrjVcZ09LS9FyJ9PAd5zpmb28Pe3v7Zp+3sLBockldW3369Gnzazx+57aTk1ObX6+15HK5+KcoLZHafmPth89JMcYkjUOKMSZpHFKMMUnjkGKMSRqHFGNM0jikGGOSxiHFGJM0DinGmKRxSDHGJI1DijEmaRxSjDFJ45BijEkahxRjTNI4pBhjksYf1dKCHTt26LuEDuHYsWP6LqHDKy4uhoODg77LkCT+0Ds1Gj/0jrH2FBwczB96pwaHFNOptn5SJ2NP4nNSjDFJ45BijEkahxRjTNI4pBhjksYhxRiTNA4pxpikcUgxxiSNQ4oxJmkcUowxSeOQYoxJGocUY0zSOKQYY5LGIcUYkzQOKcaYpHFIMcYkjUOKMSZpHFKMMUnjkGKMSRqHFGNM0jikGGOSxiHFGJM0DinGmKRxSDHGJI1DijEmaRxSjDFJ45BijEkahxRjTNI4pBhjksYhxRiTNA4pxpikcUgxxiSNQ4oxJmnG+i6AdVzZ2dmoqalp0v7bb7+hW7duKm0jRoyAtbV1e5XGDIhARKTvIljHNHXqVKSlpT21n4WFBUpLSyGXy9uhKmZoeLrHtDZ9+vSn9jE2Nsaf/vQnDiimNQ4pprWJEyeic+fOLfZRKpUIDw9vp4qYIeKQYlozNTVFcHAwTE1Nm+3TuXNnTJgwoR2rYoaGQ4q1yYwZM1BXV6f2ORMTE0ybNq3FEGPsafjEOWuThoYG2Nra4s6dO2qf//HHHzFu3Lh2rooZEj6SYm1iZGSEsLAwtUdLPXv2hKenpx6qYoaEQ4q12fTp05tM+UxNTREVFQUjIx5irG14usd0wtHREVevXlVpO3XqFF5++WU9VcQMBf+aYzoREREBExMT8XHfvn05oJhOcEgxnQgLC4NCoQDwaKoXHR2t54qYoeDpHtOZwYMHIzc3FwBw4cIFDBgwQM8VMUPAR1JMZyIjIwEAw4YN44BiOsMhxXRm2rRpEAQBERER+i6FGRLSQmpqKgHgH/7hH/7R+Cc4OFibuKE2fZ5UampqWxZnBqiwsBDLly9HTEwM3N3d9V0Ok4iNGzdqvWybQmrq1KltWZwZqOXLl8Pd3Z3HBxNp8rljzeFzUowxSeOQYoxJGocUY0zSOKQYY5LGIcUYkzQOKcaYpHFIMcYkjUOKMSZpHFKMMUnjkGKMSRqHFGNM0jikGGOSxiHFGJM0DinGmKRxSDHGJK1NnyfFtFNfX4/du3fj888/x5tvvomFCxcCAMrLyzF27FgsW7YMUVFROl/v1atXkZKSgpKSEgwfPhxhYWEqX0Olrv+XX36JlJQUXLlyRef1AEBWVhaKiooAAJ06dUJgYGCL/S9fvoycnBwAj7492c/PD127dm1zHdnZ2SrbKAgCbGxs4OTkhL59+7a4n7QhxTFw9OhRFBYWqvQ3NjaGtbU1unXrhqFDh6JTp046r+lpOKT0oLi4GDdu3MDhw4fh6+srthsbG6N79+6wtLTU+Tpzc3MxcuRIdO3aFbdv34ZCoUB8fDyysrKaXd/ly5fx448/ori4WOf1NHJ3d8ft27cRGhoKADhx4gRGjBjRbP8///nPyMjIgIeHB7Zt26aTgAKAsWPH4v79+/D390ePHj2wbNky5OfnY+HChSgvL0dycjJ8fHx0si5AmmNgzJgxePDgAfz8/GBtbY13330XTk5OuHz5Mk6ePIl9+/Zh/Pjx+PjjjzFw4ECd19estnzGOdPenTt3CABt2LChXda3ePFiOnbsGBERFRcXU2hoKAGg5cuXt7jc0qVLSSaTtWpdACg1NVXj/vX19WRtbU0AKCgoqNl+Fy9eJLlcTgBo1apVrapJU1ZWVjR48GDxcVlZGTk7O5MgCHT69GmdrkuqY6Bbt240YMCAJssfOnSI7OzsyNzcnI4fP96qdQcHB2v9Ged8TkpPjI3b7yC2oqICHh4eGD16NACgd+/eWL9+PQRBwM8//9zisrqe5qgjk8nQr18/+Pj4ID09HXl5eWr7ffTRR+LXZj2LIw3g0RebCoIgPu7WrRtCQ0NBRNi+fbtO1yXVMWBqaqr2Nby9vbF582bU1NQgKCgItbW1z7xuoB2ne1euXEFiYiKWL1+O27dvIykpCba2tpg+fTq6dOmCgoICpKWlid9+++RhfF5eHr7//ntUVFRg5MiR8PPzAwAcPnwYJSUlYj9/f39cunQJ+fn5AAAfHx90795doxoLCgrw3XffISYmBtnZ2di7dy9cXFwQEREBI6P/5nltbS2ysrKQlZUFe3t7+Pr6ol+/fiqv9bQ+j78RGtXU1CAtLQ22trbi1CI/Px9JSdk3dqMAACAASURBVElYs2YNCgoKsGPHDtjY2GDmzJkqAfLw4UN8/fXXKCkpwaBBg+Dt7Q0rKysYGRmhS5cuCAgIUFlX3759MXjwYPTv31+lXaFQID09HadPn4aXlxcaGho02ne6EBsbiwMHDuAf//gHvvjiC5XnSkpKcOrUKcTFxeHzzz9Xu/yzGiNWVlYAoDLtNeQx0BJ/f394e3vj8OHDSEtLQ3h4uMbLak2bw6/WTve2bdtGDg4OBIDS0tIoIiKCwsLCSCaTUWBgIGVlZdG0adMoLCyMjI2Nyd/fX2X5BQsWkKenJ925c4cOHDhAgiDQBx98QEREd+/epejoaAJA4eHhRERUVFREFhYWlJmZSQ0NDRrVGB8fT5aWltSrVy9KSUmhoUOHkoWFRZMpSHV1NXl5edH27dupvLyc4uPjqXPnzvTNN9+0qk9lZaXKof758+dpypQpBIDWr19PRERJSUlka2tLAOjbb7+lwMBAmjhxIgGglStXiq9VWlpK/fr1o+TkZKqrq6PY2FgCQI6OjuTh4aF2e5VKJcnlcpWaKioqyNvbm+Li4qisrIySk5PJ1NT0mU/3iIhefvllIiJyc3MjMzMzunnzpsrzK1eupMTERMrMzFQ7RdLVGOnRowcNGTJE5bWHDx9OAGjLli1EZNhjgIjIzs5O7XSv0XvvvUcAaNasWc32eVJbpnvtdk5q9erVBIB2794tts2dO5cA0NatW8W2xh1QWVkptllbW9PatWvFx66urjR69GjxcV1dHXl4eFDnzp3p2rVrtHDhQtq1a1ertys0NJTkcjn9+9//JiKiGzdukLu7OwGg/fv3ExHRjBkzKDo6WmW54OBgsrCwoKKiIo37PDlAiYiuX7+uMkCJSBxsj++38ePHk4uLi/g4JiaGrKysSKFQENGjNyAA+stf/tLstqanp9OoUaNU3qBz586lKVOmqPSbNGlSu4bUtm3bCACtWLFCfK6qqoqGDRtGtbW1zYaUrsZIjx49qG/fvnTy5EnKzs6madOmEQCaOXOmuK8MeQwQPT2ktm7dSgBowoQJzfZ5Uoc4J9V4DuG1114T21566SUAgIeHh9jWeNXg+vXrYtuePXswZ84cAI+u/hARqqurxedNTEywdetWAMCbb74Jc3PzJoe2mpDL5bCyskJYWBgAoFevXli3bh0A4ODBg3j48CHS0tLg5uamstycOXNQXV2NxMREjfo0R915FrlcDuDRYXajIUOGqEw98vLyYGRkJE4fHBwc4OzsjOzsbLXrUSgUWLduHZKTk8VlSkpKkJCQ0OQK1rBhw5qt91kICQmBo6MjPv/8c9y/fx8AsGXLFoSHhzd7rgTQ7RiRyWS4evUqTp48CR8fH5w5cwaJiYkQBMGgx4CmqqqqAAA9e/Zs1XLaareQajyn8/gOMTMza9KvcSAqFAqxbezYscjKykJERAQuXboER0dHEJHKck5OTvj73/+Os2fPwtXVVes6n/wPa7wcXlRUhJycHCgUiiYnPBvn9JcuXdKoT3MeP+/VUptcLkd9fb342MPDAxUVFThx4gSAR+dCbty4gZdfflntemJiYrBq1SoMGDBAbDt79iwUCgXs7OxU+rZ2ALeVTCbD4sWLUVFRgS+++AJKpRKbN2/G22+/3eJyuhwjnTp1QlBQEGJiYhAdHS3+MgVg0GNAUxcvXgSANr3PWqNDXN2LjY3Fli1bkJCQgPDwcLXh1tDQgOzsbHh7e+Pdd98VbxBsK1NTU5iZmeGFF16AUqkEAPFmwkY9evQAALi4uGjUR9cWLVqE4OBgxMbG4tChQ1iyZAnGjBmDNWvWNOn7ySefYMSIESq/lQGIRy03b97UeX2tNWvWLHTr1g3//Oc/sX37drz++uviyevmtNcYMeQxoIm6ujpkZmbC2NhYq9mKNiQfUqdOncKGDRswb948mJubi+1P/pZcu3YtoqKikJKSAhMTE0RFRTXpo4mamhqVxzk5OaitrcXIkSPh5uYGMzMzHD16VKVPaWkpAMDT01OjPromCALs7e2xceNGNDQ0YN68eTh48CA6d+6s0m/Lli0QBAEzZ84U24gIFy5cEKfZe/fubfL67XmFD3h0lDBv3jxcv34dCxYsQExMTIv9dTlG6NF52mbXZchjQBMbNmxAfn4+YmJiDO9I6sGDBwD+O58F/jule/zcQeMhbGO/xtvwMzIyUF9fj0OHDuHs2bMoLy9HXl4eCgsLsW/fPpSVlcHPzw+2trb48MMPceTIEXz44YetrrOyshLXrl0TH+/btw+vvvoqgoKCYGNjgwULFqCwsBBHjhwR+2RkZCAkJATjxo3TqE9z+0Nd2927d9XuI4VCId6n8uGHH4p/XmJiYoLKykrk5uaqTAc2bdqEr776ClZWVkhKSkJiYiLi4+MxadIklJaWwtXVFb6+vsjMzERSUhKAR781z5w5AyJCUVGRyuvp0sOHD1XOQQLAggULYG5ujsmTJ6N3795ie+ORXuORH6C7MVJXV4fy8nLcu3ev2VoNeQwAj96Tjf9+XG1tLRYtWoTVq1djxYoVWLt2bbP7SOe0Odve2qt7Bw4coCFDhhAAeuedd+jixYv0/fff04gRI8TLwufOnaMffviBPDw8CAAFBwdTbm4uERFFRESQkZER2dra0qZNm2jt2rVkZGRES5YsoZ07d1Lnzp1p1qxZVF9fT0RE//rXvwgAyWQyWr58OT18+FCjOmfNmkVyuZwmT55Mn332Gc2ePZs8PDyosLBQ7KNUKmnx4sXUs2dPWrZsGUVFRdHUqVOpurpa4z43btygOXPmEABydXWljIwMunbtmkrb3r17KSMjgxwdHQkALVy4kC5fvkzbt28nJycnAkBLly6l27dv03fffUfm5uYEQOXnhRdeoAMHDlBiYmKT5xp/nJycxKs7t27dIk9PTwJALi4uNHnyZAoPDydLS0uaP38+FRcXa7Qf0Yqrez/++CMFBgaKY+Pnn38Wn5szZw799ttvRPTorvTPPvuMBgwYQADIwcGBPvroI7p79y4RtX2M7Nu3T6wDAM2ZM4dOnDihtmZDHQP/+c9/KCgoiACQsbExubm5UUBAAAUFBdGkSZPonXfeoVOnTmn0//qktlzdE4haPyfasWOHeBdueyktLUWXLl3Em9fKy8t19ndbjf7nf/4H+/btQ2FhIXJzc2FtbQ0nJye1faurq3HhwgUMGjRIZYrR2j66sHPnTtTX12PChAkoKytDVVUVKisr8fvvv+Obb75R+W2uiYKCAiiVSvTv3x9XrlxB9+7dn3pO6HGCICA1NRVTp05t7aa0SXuMkccZ8hjQtZCQEABAWlpaq5ftMH9g/OTlzmc5+ExNTTF8+PAW+1hYWDS5xKxNn7bKz88Xz980/nFqI1dXV/FqT2s8fld0cyEtRe05RgDDHgNS0mFCqj08fPhQ5VxAR1BUVISSkhJERERgzpw5YqhcvHgRX331Fd5//309V8ieNUMfAwYfUkVFRYiOjm6xj0KhQF1dHa5evYr79+9j5cqVePvtt+Hg4NBOVWpv/Pjx2L9/PzIzM/HOO++gsLAQ/fv3xxtvvIEvv/yyVdM01jEZ+hgw+JBycHDAnj17ntrP2NgYMpmsHSrSPR8fH/FOcSJq9xswmf4Z8hgw+JASBEHtjX2GypAGJ9OOoY0Byd/MyRh7vnFIMcYkjUOKMSZpHFKMMUnjkGKMSRqHFGNM0jikGGOSxiHFGJM0DinGmKRxSDHGJI1DijEmaW362z1D+xshpjuhoaEIDQ3VdxlMQoKDg7VaTqtP5iwuLm7yTRiMAY/CKSYmBu7u7vouhUlMnz59tBoXWoUUY83R10cHM8PF56QYY5LGIcUYkzQOKcaYpHFIMcYkjUOKMSZpHFKMMUnjkGKMSRqHFGNM0jikGGOSxiHFGJM0DinGmKRxSDHGJI1DijEmaRxSjDFJ45BijEkahxRjTNI4pBhjksYhxRiTNA4pxpikcUgxxiSNQ4oxJmkcUowxSeOQYoxJGocUY0zSOKQYY5LGIcUYkzQOKcaYpHFIMcYkjUOKMSZpHFKMMUnjkGKMSRqHFGNM0gQiIn0XwTomX19f5OTk4PEh9ODBA5ibm8PY2FhsMzExwe+//w57e3t9lMk6OD6SYlrz8/PD/fv38eDBA/EHAGpqasTHVVVVcHFx4YBiWuOQYlqbPn06jIxaHkIymQxRUVHtVBEzRBxSTGs2NjZ47bXXIJPJmu1DRAgKCmrHqpih4ZBibRIREdHsczKZDN7e3rCxsWnHipih4ZBibRIcHNzslI+IEB4e3s4VMUPDIcXaxMrKCn5+fipX8xoZGxvjT3/6kx6qYoaEQ4q1WXh4OJRKpUqbsbExJk+eDCsrKz1VxQwFhxRrszfffBOdOnVSaVMqlQgLC9NTRcyQcEixNjM3N0dgYCBMTEzENgsLC/j6+uqxKmYoOKSYTsyYMQMKhQLAozvMp02bBnNzcz1XxQwBhxTTiddffx3dunUDACgUCsyYMUPPFTFDwSHFdMLY2BjTp08HAHTv3h1eXl76LYgZDA4ppjPTpk0DAISFhbV4FzpjrdH05hYDd+zYMXz88cf6LsNgderUCb///jtCQkL0XYpBcnd3x+LFi/VdRrt67o6kioqKsHPnTn2XYbAGDhwonptqtHPnThQXF+upIsNx/PhxHDt2TN9ltLvn7kiqUVpamr5LMEhEBEEQVNoEQcCiRYswdepUPVVlGJ7Xo9Pn7kiKPVtPBhRjbcUhxRiTNA4pxpikcUgxxiSNQ4oxJmkcUowxSeOQYoxJGocUY0zSOKQYY5LGIcUYkzQOKcaYpHFIMcYkjUOKMSZpHFKMMUnjkGKMSRqHFGNM0jikOpjc3FxYWlqiT58+MDU1hSAIeOWVV/DgwYNmlzly5AjGjh2r9qvQn+bQoUN44403IAgCVq9e3ZbSW23dunXo06cP1q9fj7Nnz6KsrAz79++Hl5cXbGxskJGR0a71AE/f/0SE/fv3w8jICF27dsWqVauQlJSEv//97wgICICFhQX8/f1x4cKFdq+9w6LnTGpqKnXkzV68eDEdO3aMiIiKi4spNDSUANDy5ctbXG7p0qUkk8lavb7q6moqLCwkABQXF6dVzQAoNTW1Vcts3LiRTExMKDs7W21No0aNoi1btmhVT1touv+7detGAwYMaLL8oUOHyM7OjszNzen48eOtWndwcDAFBwdrX3wHxUdSHUhFRQU8PDwwevRoAEDv3r2xfv16CIKAn3/+ucVlH/924dYwNzdH7969tVpWW9u2bcOiRYvwt7/9DWPHjlVbU3x8PO7evduudbVm/5uamqp9DW9vb2zevBk1NTUICgpCbW3tM6+7o3tuP+O8tWpqapCWloY//vgD7u7u8PHxgYWFhfh8bW0tsrKykJWVBXt7e/j6+qJfv37i8/n5+UhKSsKaNWtQUFCAHTt2wMbGBjNnzoSJiQn27NmDe/fuAXj0EbwBAQEwMzPDr7/+iosXLwIAJk6ciICAAJW6+vbti8GDB6N///4q7QqFAunp6Th9+jS8vLzQ0NCg9ba399dTrV27FgDg7+/fbJ8RI0aAiMTHUtv/LfH394e3tzcOHz6MtLQ0hIeHa7zsc0nfh3LtTZvp3pUrV8jT05MSEhKoqKiIvL296cUXX6Tq6moiejT98PLyou3bt1N5eTnFx8dT586d6ZtvviEioqSkJLK1tSUA9O2331JgYCBNnDiRANDKlSuJiOjmzZv0yiuvEACVKU5DQwP5+/vTtm3b1NamVCpJLpeL6yIiqqioIG9vb4qLi6OysjJKTk4mU1NTraZ7jTUAoNWrV2u1PFox3cvLyyMAZGxsTLW1tRotI7X9T0RkZ2endrrX6L333iMANGvWLI22kej5ne5xSGlgwoQJNHv2bPFxZmYmCYJAu3btIiKiGTNmUHR0tMoywcHBZGFhQUVFRUREFBsbSwBo9+7dYp/x48eTi4uL+PjgwYMEgFJSUsS22tpaCgwMbLa29PR0GjVqFDU0NIhtc+fOpSlTpqj0mzRpUocIqcZ9MHDgQI1fX2r7n+jpIbV161YCQBMmTNBsI+n5DSk+J/UU+fn5OHjwIKZMmSK2TZw4Ebdu3UJAQAAePnyItLQ0uLm5qSw3Z84cVFdXIzExEQAgl8sBqE5hhgwZovJ9dK+//joGDRqETZs2iW27du1q9quMFAoF1q1bh+TkZPFbWkpKSpCQkAAfHx+VvsOGDdNm89tda69ASm3/a6qqqgoA0LNnz1Yt9zzikHqK8+fPAwAsLS1V2m1sbAAAOTk5UCgUTd5cjecoLl26BAAwMmq6q+VyOerr61Xa5s+fj59++gnnzp0DAKSnpyMwMFBtbTExMVi1ahUGDBggtp09exYKhQJ2dnYqfTvKV0017re8vDxUV1c/tb/U9r+mGs9zubq6tnrZ5w2H1FM0nhzfv39/k+dKS0uhVCoBPHqzPK5Hjx4AABcXl1atLzIyElZWVvj0009x/vx59O/fX+2Vok8++QQjRoxocnL5/v37AICbN2+2ar1S0bt3bwwePBhKpRK///57i32JSHL7XxN1dXXIzMyEsbFxkxPxrCkOqadwdXWFkZERvvvuO/ENAQAFBQX45Zdf4ObmBjMzMxw9elRludLSUgCAp6dnq9ZnaWmJqKgopKSkYMOGDZg9e3aTPlu2bIEgCJg5c6bYRkS4cOECBg4cCADYu3dvk+W0vcJH//8qGj12Ne1ZarxpNDY2FnV1dWr7lJaWIjExUXL7XxMbNmxAfn4+YmJi+EhKAxxST2Fvb4/IyEicO3cOISEh+OGHH/DZZ59h5cqV8PX1hY2NDRYsWIDCwkIcOXJEXC4jIwMhISEYN24cAIj39Dw+hamvr4dCoWhyr8y8efNQXV2NsrIyvPDCCyrPbdq0CV999RWsrKyQlJSExMRExMfHY9KkSSgtLYWrqyt8fX2RmZmJpKQkAI9+c585cwZEhKKioiZTnKdpDIrG8yjPWlBQENauXYusrCy89dZbTdZ77do1fPDBB4iIiJDc/gcenatq/PfjamtrsWjRIqxevRorVqwQb7VgT6HPs/b6oM3VvcrKSgoICCAABIAcHR3p5MmT4vNKpZIWL15MPXv2pGXLllFUVBRNnTpVvEUhIyODHB0dCQAtXLiQLl++TNu3bycnJycCQEuXLqXbt2+rrNPHx4f27dun0paYmCjW8OSPk5OTeIXp1q1b5OnpSQDIxcWFJk+eTOHh4WRpaUnz58+n4uJijbc9JyeH5s6dSwDI2dmZPvvsM1IoFK3af9DijnMiop07d9KQIUPIwsKCxo0bR/PmzaO33nqL4uLiVG5PkNL+/89//kNBQUHibRRubm4UEBBAQUFBNGnSJHrnnXfo1KlTrd4XRM/v1T2BqJ2O4SVix44dCA0N1WrqcuPGDfFoRd0d3NXV1bhw4QIGDRoEc3PzNtVZVFQEBweHNp3wLigogFKpRP/+/XHlyhV0794dVlZWbapLG4IgIDU1FVOnTtVq+aqqKvz2229wdnYWzzWpI7X9r2uNVxnT0tL0XEn74jvOW8He3h729vbNPm9hYdHkUri2+vTp0+bXePyOaycnpza/nr7I5XLxT1FaIrX9z3SDz0kxxiSNj6SeM0VFRYiOjn5qv6ioKERERLRDRYy1jEPqOePg4IA9e/Y8tZ82nz3F2LPAI/E5IwgCzMzM9F0GYxrjc1KMMUnjkGKMSRqHFGNM0jikGGOSxiHFGJM0DinGmKRxSDHGJI1DijEmaRxSjDFJ45BijEkahxRjTNKe27/da+5riljblJeXQy6XN/nygo0bNz53H9ama8ePH9foc7UMzXN3JNWnTx8EBwfruwyDdfjwYZSUlKi0BQcHw8HBQU8VGY7Ro0fD3d1d32W0u+fu44PZs9XWjwpm7EnP3ZEUY6xj4ZBijEkahxRjTNI4pBhjksYhxRiTNA4pxpikcUgxxiSNQ4oxJmkcUowxSeOQYoxJGocUY0zSOKQYY5LGIcUYkzQOKcaYpHFIMcYkjUOKMSZpHFKMMUnjkGKMSRqHFGNM0jikGGOSxiHFGJM0DinGmKRxSDHGJI1DijEmaRxSjDFJ45BijEkahxRjTNI4pBhjksYhxRiTNA4pxpikcUgxxiSNQ4oxJmnG+i6AdVyrV6/GjRs3mrQnJCTg8OHDKm1xcXHo1atXe5XGDIhARKTvIljHtGTJEnz00UcwMTER2xoaGiAIAgRBAAAolUrY2dmhuLhYbGOsNXi6x7Q2ffp0AIBCoRB/lEol6uvrxccymQxRUVEcUExrfCTF2sTZ2RkFBQUt9jl37hyGDh3aThUxQ8NHUqxNwsPDVaZ7T3J2duaAYm3CIcXaJDw8HAqFQu1zJiYmiI6ObueKmKHh6R5rs5deegm//fYb1A2lvLw8ODs766EqZij4SIq1WWRkJGQymUqbIAh45ZVXOKBYm3FIsTabPn06GhoaVNpkMhkiIyP1VBEzJBxSrM3s7e0xZswYGBn9dzgplUoEBwfrsSpmKDikmE5ERESI/5bJZPDy8oK9vb0eK2KGgkOK6URISIjKkdTjocVYW3BIMZ3o2rUrfHx8IAgCjIyMEBAQoO+SmIHgkGI6Ex4eDiKCn58funTpou9ymIFo8ikIxcXFyMnJ0UctrIOrr6+HmZkZXnzxRezYsUPf5bAOqE+fPnB3d1dtpCekpqYSAP7hH/7hn3b/CQ4OfjKSqNnPk+Ib0Zk2KisrYW1t/dR+giAgNTUVU6dObYeqWEcQEhKitp3PSTGd0iSgGGsNDinGmKRxSDHGJI1DijEmaRxSjDFJ45BijEkahxRjTNI4pBhjksYhxRiTNA4pxpikcUgxxiSNQ4oxJmkcUowxSeOQYoxJGocUY0zSOKQYY5LW7IfeMe3U19dj9+7d+Pzzz/Hmm29i4cKFAIDy8nKMHTsWy5YtQ1RUlM7Xe/XqVaSkpKCkpATDhw9HWFgYTExMWuz/5ZdfIiUlBVeuXNFqnQcOHIBCocDEiRO1rFozWVlZKCoqAgB06tQJgYGBLfa/fPmy+BHYRkZG8PPzQ9euXdtcR3Z2tsq+EgQBNjY2cHJyQt++fVvc39qQ4lg6evQoCgsLVfobGxvD2toa3bp1w9ChQ9GpUyed1sNHUjpWXFyMGzdu4PDhw1AoFGK7sbExunfvDktLS52vMzc3F4MHD8bnn3+Of/3rX4iOjsbo0aPx4MGDZpe5fPkyfvzxRxQXF7d6fYcOHcIbb7yBN954A7/88ktbSteIu7s7TE1NERERgaCgIJw8ebLF/n/+858RERGBL774Aq+99ppOAgoAxo4di+7duyMiIgKLFi3CzZs3kZOTg8mTJ+OFF17AgQMHdLKeRlIcS2PGjEHPnj0RGRmJBQsWIC8vDzU1NTh9+jQ++OADdO/eHf7+/rhw4YLuimruM86Z9u7cuUMAaMOGDe2yvsWLF9OxY8eIiKi4uJhCQ0MJAC1fvrzF5ZYuXUoymazV66uurqbCwkICQHFxcVrVDIBSU1M17l9fX0/W1tYEgIKCgprtd/HiRZLL5QSAVq1apVVtT2NlZUWDBw8WH5eVlZGzszMJgkCnT5/W6bqkOpa6detGAwYMaLL8oUOHyM7OjszNzen48eOtWndwcLDazzjnI6lnwNi4/WbRFRUV8PDwwOjRowEAvXv3xvr16yEIAn7++ecWl9V2emJubo7evXtrtay2ZDIZ+vXrBx8fH6SnpyMvL09tv48++giRkZEA8EyONADA1NQUgiCIj7t164bQ0FAQEbZv367TdUl1LJmamqp9DW9vb2zevBk1NTUICgpCbW1tm+vSyR64cuUKEhMTsXz5cty+fRtJSUmwtbXF9OnT0aVLFxQUFCAtLQ2mpqaIjo5ucvidl5eH77//HhUVFRg5ciT8/PwAAIcPH0ZJSYnYz9/fH5cuXUJ+fj4AwMfHB927d9eoxoKCAnz33XeIiYlBdnY29u7dCxcXF0RERKh8825tbS2ysrKQlZUFe3t7+Pr6ol+/fiqv9bQ+jw/gRjU1NUhLS4OtrS18fHwAAPn5+UhKSsKaNWtQUFCAHTt2wMbGBjNnzlQJkIcPH+Lrr79GSUkJBg0aBG9vb1hZWcHIyAhdunRp8kWcffv2xeDBg9G/f3+VdoVCgfT0dJw+fRpeXl5oaGjQaN+pI5PJtF62LWJjY3HgwAH84x//wBdffKHyXElJCU6dOoW4uDh8/vnnapd/VmPNysoKAFSmz4Y8llri7+8Pb29vHD58GGlpaQgPD9d4WbWePLRq7XRv27Zt5ODgQAAoLS2NIiIiKCwsjGQyGQUGBlJWVhZNmzaNwsLCyNjYmPz9/VWWX7BgAXl6etKdO3fowIEDJAgCffDBB0REdPfuXYqOjiYAFB4eTkRERUVFZGFhQZmZmdTQ0KBRjfHx8WRpaUm9evWilJQUGjp0KFlYWDSZOlRXV5OXlxdt376dysvLKT4+njp37kzffPNNq/pUVlaqHKKfP3+epkyZQgBo/fr1RESUlJREtra2BIC+/fZbCgwMpIkTJxIAWrlypfhapaWl1K9fP0pOTqa6ujqKjY0lAOTo6EgeHh5qt1epVJJcLlepqaKigry9vSkuLo7KysooOTmZTE1NtZruERE1NDQQAFq9erVWy6OV0z0iopdffpmIiNzc3MjMzIxu3ryp8vzKlSspMTGRMjMz1U6RdDXWevToQUOGDFF57eHDhxMA2rJlCxEZ9lgiIrKzs1M73Wv03nvvEQCaNWtWs32e1Nx0TyfnpFavXk0AaPfu3WLb3LlzCQBt3bq1SeGVlZVim7W1Na1du1Z87OrqSqNHjxYf19XVkYeHB3Xu3JmuXbtGCxcupF27drWqPiKi0NBQksvl9O9//5uIiG7cuEHu7u4EgPbv309ERDNmzKDo6GiV5YKDg8nCwoKKioo07vPkkf6SjQAAIABJREFUwCIiun79usrAIiJxkDy+38aPH08uLi7i45iYGLKysiKFQkFEj944AOgvf/lLs9uanp5Oo0aNUnljzZ07l6ZMmaLSb9KkSR0ypLZt20YAaMWKFeJzVVVVNGzYMKqtrW02pHQ11nr06EF9+/alkydPUnZ2Nk2bNo0A0MyZM8V9bshjiejpIbV161YCQBMmTGi2z5Oe6Tmpxrn/a6+9Jra99NJLAAAPDw+xbeDAgQCA69evi2179uzBnDlzAAAnTpwAEaG6ulp83sTEBFu3bgUAvPnmmzA3N29ySKoJuVwOKysrhIWFAQB69eqFdevWAQAOHjyIhw8fIi0tDW5ubirLzZkzB9XV1UhMTNSoT3PUnR+Ry+UAHh0eNxoyZIjKlCEvLw9GRkbiYb+DgwOcnZ2RnZ2tdj0KhQLr1q1DcnKyuExJSQkSEhLEqUGjYcOGNVuvlIWEhMDR0RGff/457t+/DwDYsmULwsPDmz1XAuh2rMlkMly9ehUnT56Ej48Pzpw5g8TERAiCYNBjSVNVVVUAgJ49e7ZqOXV0ElKN53Qe3xAzM7Mm/RoH0OOXU8eOHYusrCxERETg0qVLcHR0bPLFpE5OTvj73/+Os2fPwtXVVes6n9zRI0aMAAAUFRUhJycHCoWiyYnKxrn4pUuXNOrTnMfPe7XUJpfLUV9fLz728PBARUUFTpw4AeDROYwbN27g5ZdfVruemJgYrFq1CgMGDBDbzp49C4VCATs7O5W+rR14UiGTybB48WJUVFTgiy++gFKpxObNm/H222+3uJwux1qnTp0QFBSEmJgYREdHi7+UARj0WNLUxYsXAaBN79dGer+6Fxsbiy1btiAhIQHh4eFqw62hoQHZ2dnw9vbGu+++K97Y11ampqYwMzPDCy+8AKVSCQDiTYCNevToAQBwcXHRqI+uLVq0CMHBwYiNjcWhQ4ewZMkSjBkzBmvWrGnS95NPPsGIESNUfpsCEI82bt68qfP69GXWrFno1q0b/vnPf2L79u14/fXXxZPXzWmvsWbIY0kTdXV1yMzMhLGxsVaznifpNaROnTqFDRs2YN68eTA3Nxfbn/zttnbtWkRFRSElJQUmJiaIiorS6mvga2pqVB7n5OSgtrYWI0eOhJubG8zMzHD06FGVPqWlpQAAT09PjfromiAIsLe3x8aNG9HQ0IB58+bh4MGD6Ny5s0q/LVu2QBAEzJw5U2wjIly4cEGcZu/du7fJ62t7ha9x/2vz/6ALcrkc8+bNw/Xr17FgwQLExMS02F+XY40encttdl2GPJY0sWHDBuTn5yMmJkY6R1KNd6M2zkOB/07pHp/zNx56NvZrvH0+IyMD9fX1OHToEM6ePYvy8nLk5eWhsLAQ+/btQ1lZGfz8/GBra4sPP/wQR44cwYcfftjqOisrK3Ht2jXx8b59+/Dqq68iKCgINjY2WLBgAQoLC3HkyBGxT0ZGBkJCQjBu3DiN+jS3P9S13b17V+0+UigU4v0lH374ofhnISYmJqj8f+3de1RUV5o28KeoAkFuCuKtNYIXiHh3TKJGowaJisaoIIgWIPYa2+ugHSVxGaNGs1Yyro4TJ2ZUkhbTQxSwW01r04ltD217zRo0RlskCN6YeEHjjUKoAt7vDz+qrQBaBUXVBp7fWqxY++za9bJzeDhnn+LU/fs4f/68xWH8li1b8Nlnn8HHxwepqanYvn07/vM//xOTJk1CcXExQkNDMX78eOzfvx+pqakAHv+2++677yAiuHbtmsV41jAajTW+n8ZUWlpqsZYJAIsXL4a7uzsmT55s8b6t6iPG6iNIwH77mtFoxN27d/HgwYM6a23O+xLw+Ge7+t9PKi8vx9KlS7F27VqsWLEC69evr3OObPLzlXRbr+5988030rdvXwEg8+bNk7y8PPnTn/4kL7zwgvly7vfffy9//etfZcSIEQJAoqKi5Pz58yIiEhcXJy4uLtKhQwfZsmWLrF+/XlxcXGTZsmWye/du8fb2ljlz5khFRYWIiHz66acCQLRarbz99ttSWlpqVZ1z5swRT09PmTx5smzevFnmzp0rI0aMkEuXLpn7VFZWyq9//WsJCAiQt956SxISEiQ6OloePXpkdZ8ff/xR5s+fLwAkNDRU9u7dK1evXrVoy8rKkr1790pgYKAAkKSkJCksLJRdu3ZJUFCQAJDly5fLzZs35Y9//KO4u7sLAIuv5557Tr755hvZvn17jW3VX0FBQearMjdu3JCRI0cKAAkODpbJkyeLXq8XLy8vWbRokRQVFVn9//zYsWPmq7c9e/aUzZs3m68YWQs2XN3Lzs6WadOmmfexkydPmrfNnz9fzp49KyKP35W+efNmCQkJEQDSpUsX+c1vfiM//fSTiDR8X/vzn/9srgOAzJ8/X7799ttaa26u+9Lhw4clMjJSAIhOp5NBgwbJ1KlTJTIyUiZNmiTz5s2TnJwcm/aFanVd3dOIWB63ZmRkmN896yjFxcVo06aN+U1nd+/etdvfW1X75S9/iT//+c+4dOkSzp8/D19fXwQFBdXa99GjR7hw4QJ69+5tcWpgax972L17NyoqKhAeHo47d+7AYDDg/v37OHfuHH7/+99b/Ba2RkFBASorK9GrVy9cvnwZ/v7+z1zLaQwajQbp6emIjo526Os6Yl97UnPel+xt+vTpAIDMzEyLdiXugvDzy5SNudO4ublh4MCBT+3j4eFR49Jwffo01MWLF83rLtV/VFotNDTUfJXGFk++m7mukG7OHLmvAc17X3IUJULKEUpLSx22fmIv165dw61btxAXF4f58+ebQyUvLw+fffYZ3n//fSdXSE1FU96XmnRIXbt2DYmJiU/tYzKZYDQaceXKFTx8+BCrVq3Cr371K3Tp0sVBVdbfmDFj8PXXX2P//v2YN28eLl26hF69emHcuHHYtm2b3U7TrJlHAEhISEBcXJxdXpMcy1H7UmNo0iHVpUsXHDhw4Jn9dDqd0/4gtqFee+018zvFRaRR3oBpyzxS0+WIfakxNOm9TqPR1PqGvOaqsXaqljaP1LT+2sDp7zgnInoahhQRKY0hRURKY0gRkdIYUkSkNIYUESmNIUVESmNIEZHSGFJEpDSGFBEprc4/i8nIyHBkHdQCHT9+3NklkEKKiopq/8P/n98Fr/rOnPziF7/45egvq+7MSdQQzrrjJjVfXJMiIqUxpIhIaQwpIlIaQ4qIlMaQIiKlMaSISGkMKSJSGkOKiJTGkCIipTGkiEhpDCkiUhpDioiUxpAiIqUxpIhIaQwpIlIaQ4qIlMaQIiKlMaSISGkMKSJSGkOKiJTGkCIipTGkiEhpDCkiUhpDioiUxpAiIqUxpIhIaQwpIlIaQ4qIlMaQIiKlMaSISGkMKSJSGkOKiJSmc3YB1HQdOXIEZWVlNdrPnj0LPz8/i7YXXngBvr6+jiqNmhGNiIizi6CmKTo6GpmZmc/s5+HhgeLiYnh6ejqgKmpueLpH9RYbG/vMPjqdDm+88QYDiuqNIUX1NnHiRHh7ez+1T2VlJfR6vYMqouaIIUX15ubmhqioKLi5udXZx9vbG+Hh4Q6sipobhhQ1yMyZM2E0Gmvd5urqihkzZjw1xIiehQvn1CBVVVXo0KEDbt++Xev27OxsjBo1ysFVUXPCIylqEBcXF8yaNavWo6WAgACMHDnSCVVRc8KQogaLjY2tccrn5uaGhIQEuLhwF6OG4eke2UVgYCCuXLli0ZaTk4PBgwc7qSJqLvhrjuwiLi4Orq6u5sfdunVjQJFdMKTILmbNmgWTyQTg8aleYmKikyui5oKne2Q3ffr0wfnz5wEAFy5cQEhIiJMrouaAR1JkN/Hx8QCA/v37M6DIbhhSZDczZsyARqNBXFycs0uhZoSne/WUkZGBmJgYZ5dBTURUVJRVd4ygmng/qQZKT093dglKiImJwZIlS9CxY0cEBQU5uxylbNy40dklNGkMqQaKjo52dglKiImJwbBhwzgfteARVMNwTYqIlMaQIiKlMaSISGkMKSJSGkOKiJTGkCIipTGkiEhpDCkiUhpDioiUxpAiIqUxpIhIaQwpIlIaQ4qIlMaQIiKlMaSISGm8n5QT3Lp1C6mpqTh58iQMBgOee+45FBYW4vXXX0dSUpKzy2t0f//732t8Rt/P9e/fH6Ghodi3bx/+67/+q8XMDdXEIykHKy0txbRp06DX65GRkYEbN24gJSUFhw4dMn8klLXKy8utalPNiBEj0LlzZ8TFxeGdd96Bl5cXvLy84O7ujjt37mDdunX48ssvUVRUhB9//LFFzQ3VxJBysL179+LKlSvo3LkztFotTpw4ga+++qpeY61cuRJVVVXPbFONRqPBq6++Cg8PD/j6+mLKlCmYMmUKoqKikJSUhOzsbBiNRgQGBmLmzJn1eo2mOjdUE0PKwU6fPg13d3fzY3d3d7zyyis2j3P27Fls2bLlmW0qe3IentSpUycsXLgQAKDT2b4i0Rzmhv6Ja1IOcv36dWRnZ+Po0aMwGAzYuXMnACA2NhYajabW5+Tn5+NPf/oT7t27hxdffBETJkwAABw9ehQzZ86EwWDArl274Orqis6dO9domz59OgDgwYMHSE9PR25uLrp3747Zs2fDy8sLAHDx4kWkpqbivffeQ0FBATIyMtC+fXvMnj3b4mPTHeXbb7/FkCFD0KNHDwDg3BCPpBxFp9PB09MTWq0WLi4u8PT0hKenZ539/+3f/g2//OUvodfrMXz4cEycOBEffvghAEBEMHLkSACAr68vfH19a20DHv8w6/V6dOvWDQkJCdi6dSsGDhyIe/fuYceOHRgxYgTef/99HDhwAG+//TaOHz+OuXPnYt26dY08IzVVVFRg3bp1qKysfGq/ljg3LZpQvaSnp0t9pi8mJkZCQkIs2u7fvy8AZMOGDeY2X19fWb9+vflxaGioDB061Px47dq1AkCqqqqe2hYeHi579uwxP87KyhIAsmrVKhERSU5OFgCyb98+c58xY8ZIcHCwTd8XAElPT7fpOW3bthVPT0955ZVX5JVXXpF27doJADEajeY+zWFuoqKiJCoqyqbn0D/xdE9RBw4cQO/evQE8PgUSETx69MimMa5fv46DBw9i4MCBOHnyJACgpKQEQ4YMQWlpKQCYj+YiIiLMz+vbt6+5f2Pr0aMH/va3vwEAjEajVZ9+3FLmhh5jSCnq5Zdfxp49e/CHP/wB48aNQ2BgIP7v//7PpjHy8/MBAMnJyWjXrl2tfVxcap7xe3p6oqKiwvaiG8jNzQ0LFiyotaYntcS5acm4JqWo5ORk/Pa3v0VKSgr0ej1atWpl8xhubm4AgFOnTtXY9vDhwwbX2BhGjRoFrVb71D4tdW5aKoaUgnJycrBhwwYsXLjQ4jK9iNToW9sic3VbSEgItFotVq9eDaPRaN5eXFyMtLS0RqjcdrV9T0/TkuaGHuPpnoPdvXsX9+/ft2grKSkBABgMBgBA69atATx+4+fYsWORnZ2NM2fOwGQyIT8/HzqdDgEBAQAe/9AajUa88MILtbbNmzcPmzdvxqhRo7B48WIYjUZ8+eWX5o/+/umnnwAAjx49gre3N4DHV9lMJhPKy8vrdZRijaqqKhgMBty7d++p/Vri3NDPOHXZvgmz9erenTt35KOPPpJWrVoJAElKSpLs7Gz58ccfZf78+QJAQkNDZe/evSIiEhcXJy4uLtKhQwfZsmWLrF+/XlxcXGTZsmUiIlJYWCgdOnSQtm3bymeffVZnm8FgkPj4eAEgAMTHx8d8RWvv3r0SGBhorqewsFB27dolQUFBAkCWL18uN2/etOr7gw1X97KzsyU6OloAiEajkTfffFO+/fbbGv2ay9zw6l7DaERsPN4mAEBGRgZiYmJsPl2xRXFxMdq0aWN+4+Ddu3fRtm1b83aTyYSKigp4eHg8tQ0Abt++jatXr6J37941ttmDRqNBeno6oqOj7T52bZrS3FS/cbT6CI1sw9M9hVWfolR78ocQAFxdXWu887m2NgBo165dnVexmiLOTcvBhXMiUhpDioiUxpAiIqUxpIhIaQwpIlIaQ4qIlMaQIiKlMaSISGkMKSJSGkOKiJTGkCIipTGkiEhpDCkiUhpDioiUxpAiIqXxflINVNcn7LZEMTExiImJcXYZSoqKinJ2CU0W78xZT0VFRTh27Jizy1BOTEwMlixZgmHDhjm7FKV07dqVc1JPDCmyK0ffRpiaP65JEZHSGFJEpDSGFBEpjSFFREpjSBGR0hhSRKQ0hhQRKY0hRURKY0gRkdIYUkSkNIYUESmNIUVESmNIEZHSGFJEpDSGFBEpjSFFREpjSBGR0hhSRKQ0hhQRKY0hRURKY0gRkdIYUkSkNIYUESmNIUVESmNIEZHSGFJEpDSGFBEpjSFFREpjSBGR0hhSRKQ0hhQRKY0hRURK04iIOLsIaprGjx+PY8eO4cldqKSkBO7u7tDpdOY2V1dXnDt3Dp07d3ZGmdTE8UiK6m3ChAl4+PAhSkpKzF8AUFZWZn5sMBgQHBzMgKJ6Y0hRvcXGxsLF5em7kFarRUJCgoMqouaIIUX11r59e7zyyivQarV19hERREZGOrAqam4YUtQgcXFxdW7TarUICwtD+/btHVgRNTcMKWqQqKioOk/5RAR6vd7BFVFzw5CiBvHx8cGECRMsruZV0+l0eOONN5xQFTUnDClqML1ej8rKSos2nU6HyZMnw8fHx0lVUXPBkKIGe/3119G6dWuLtsrKSsyaNctJFVFzwpCiBnN3d8e0adPg6upqbvPw8MD48eOdWBU1FwwpsouZM2fCZDIBePwO8xkzZsDd3d3JVVFzwJAiuxg7diz8/PwAACaTCTNnznRyRdRcMKTILnQ6HWJjYwEA/v7+GD16tHMLomaDIUV2M2PGDADArFmznvoudCJb1HxzC9Xq+PHj+Oijj5xdhvJat26Nc+fOYfr06c4uRWnDhg3Dr3/9a2eX0STwSMpK165dw+7du51dhvKef/5589rUz+3evRtFRUUOrkg9J06cwPHjx51dRpPBIykbZWZmOrsEpYkINBpNrds0Gg2WLl2K6OhoB1elFh5l2oZHUmRXdQUUUX0xpIhIaQwpIlIaQ4qIlMaQIiKlMaSISGkMKSJSGkOKiJTGkCIipTGkiEhpDCkiUhpDioiUxpAiIqUxpIhIaQwpIlIa7yflQFeuXMG2bduQlpaGy5cvO7scmxgMBuzYsQOFhYXw9fVFdHQ0QkJCGu31jh49ikuXLlm06XQ6+Pr6ws/PD/369avxWX/UPPFIyoEKCwuRnZ3d5O5OWVxcjAEDBsDHxwfr1q3Da6+9hqlTp2Lv3r2N9prDhw9HQEAA4uPjsXjxYuTn56OsrAynT5/GBx98AH9/f0RERODChQuNVgOpgUdSDjRmzBi8/PLLOHnypLNLscmHH36IXr16Qa/XAwBeeuklxMXFYfny5ZgyZUqjvKZGo8G4cePQtm1bBAQEYO3atRbbDx06BL1ej0GDBiE7OxsvvfRSo9RBzscjKQd78lN+m4qioiJcv34dImJu8/T0dMiHf7q5udXaHhYWhs8//xxlZWWIjIxEeXl5o9dCzsEjqUZmMpmwZ88enD59GqNHj0ZVVVWNPg8ePEB6ejpyc3PRvXt3zJ49G15eXgCAixcvIjU1Fe+99x4KCgqQkZGB9u3bY/bs2RaBd+TIEWRlZaFr165wcXHB3LlzrRrfGq+++irS09Px7rvvYt26daioqEBaWhqWLFnSgJlpuIiICISFheHQoUPIzMw0H+mpPp9kIyGrpKeni63Tde/ePQkLC5M1a9bInTt3ZMeOHeLm5iZardbc54cffpDXX39dvv76a/nuu++kb9++0qNHD7l7966kpqZKhw4dBIB89dVXMm3aNJk4caIAkFWrVpnHSE5OlrS0NDEYDLJz507x8vKyanxrGY1GGTNmjACQhIQEmT17tmzbts2muRARASDp6ek2Padjx44SEhJS5/Z33nlHAMicOXNEpGnMZ1RUlERFRdk0Dy0ZQ8pK9QmpBQsWyJQpUyzaJk2aZBFS4eHhsmfPHvPjrKwsix+a5ORkASD79u0z9xkzZowEBweLyOMA8ff3l7y8PPP2pKQkq8e3VmlpqTmoBg8eLDdv3rTp+SKNE1JffPGFAJDw8HARaRrzyZCyDU/3GsmtW7eQkpKCjz/+2KK9f//+yMrKAgBcv34dBw8exMCBA82L6SUlJRgyZAhKS0sBPF77AR6f2lTr27evub+rqyu8vb0xduxYbN26FRMmTMDKlSutHt9a3377LTp37ozly5djw4YNeOmll3D48GF07drV1qmxK4PBAAAICAhoUvNJ1mNINZIzZ87AZDKhY8eOFu1PfuRTfn4+ACA5ORnt2rWrdRwXl5rXNjw9PVFRUWF+/MknnyAuLg4REREYNmwYUlNTERAQYNX41jhx4gQSEhLw/fffw8fHB926dcOiRYuwYMEC/PGPf6z3uPaQl5cHAAgNDW0y80m24dW9RvLw4UMAj3/71qX6ytWpU6fqfL41Jk6ciIsXL2LJkiXIycnBkCFDkJuba7fxN2/ejBdffBE+Pj4AgIULFyI5ORlff/01bt++bfU49mY0GrF//37odDpMnTq1ycwn2YYh1Uief/55ADCf2j2p+gpfSEgItFotVq9eDaPRaN5eXFyMtLQ0q17HYDAgJSUFfn5+2LhxI7Kzs1FSUoKdO3faZXwAuHnzJrRarUXbr371K5hMJty6dcvqcextw4YN5jAJDQ1tMvNJNnL2olhTUZ+F8/HjxwsA2b59u4iIlJeXm68mXb16VUwmkyxcuFAAyNChQyUtLU22b98u4eHhcu/ePRERWbp0qQCQBw8emMddtmyZaDQaKSsrkwcPHkinTp3k0aNH5u3BwcGyadMmEZFnjm/t9+7l5SWlpaXmtt27d8uAAQOksrLS6nFg48K50WgUrVZbY+G8rKxMlixZIq6urrJixQopKyszb2sK88mFc9swpKxUn5C6ceOGjBw5UgBIcHCwTJ48WfR6vXh5ecmiRYukqKhIDAaDxMfHCwABID4+PuarR3v37pXAwEABIElJSVJYWCi7du2SoKAgASDLly+XgoIC8fDwkH79+smmTZtkzZo1kpiYKEajUUTkqeNbq6KiQlasWCEDBgyQTz75RFauXCkzZsyQwsJCm8axJaQOHz4skZGRAkB0Op0MGjRIpk6dKpGRkTJp0iSZN2+e5OTk1HheU5hPhpRtNCJPvI2Y6pSRkYGYmBjUZ7oKCgpQWVmJXr164fLly/D39zev71S7ffs2rl69it69e8PDw8PqsUUEjx49QmVlJfLz8xEcHFzrGwvrO/6TysvLcfHiRbRv3x4BAQE2P1+j0SA9PR3R0dH1en1bqDyf06dPBwBkZmba9LyWilf3HKBHjx7mfwcFBdXap127dvW6YqTRaMx3Axg8eHCd/eo7/pNatWqFPn36NGgMR2kK80nW4cI5ESmNR1It1LVr15CYmPjMfgkJCYiLi3NARUS1Y0i1UF26dMGBAwee2U+n4y5CzsU9sIXSaDRo1aqVs8sgeiauSRGR0hhSRKQ0hhQRKY0hRURKY0gRkdIYUkSkNIYUESmNIUVESmNIEZHSGFJEpDSGFBEpjX+7Z6PqG5ZR7e7evQtPT886Px5948aNLf5mbydOnMDQoUOdXUaTwSMpK3Xt2hVRUVHOLkN5hw4dqvPDGaKiotClSxcHV6SeoUOHYtiwYc4uo8ng7YPJrhx5i2BqGXgkRURKY0gRkdIYUkSkNIYUESmNIUVESmNIEZHSGFJEpDSGFBEpjSFFREpjSBGR0hhSRKQ0hhQRKY0hRURKY0gRkdIYUkSkNIYUESmNIUVESmNIEZHSGFJEpDSGFBEpjSFFREpjSBGR0hhSRKQ0hhQRKY0hRURKY0gRkdIYUkSkNIYUESmNIUVESmNIEZHSGFJEpDSGFBEpTefsAqjpWrt2LX788cca7SkpKTh06JBF25o1a9CpUydHlUbNiEZExNlFUNO0bNky/OY3v4Grq6u5raqqChqNBhqNBgBQWVmJjh07oqioyNxGZAue7lG9xcbGAgBMJpP5q7KyEhUVFebHWq0WCQkJDCiqNx5JUYP07NkTBQUFT+3z/fffo1+/fg6qiJobHklRg+j1eovTvZ/r2bMnA4oahCFFDaLX62EymWrd5urqisTERAdXRM0NT/eowQYMGICzZ8+itl0pPz8fPXv2dEJV1FzwSIoaLD4+Hlqt1qJNo9HgX/7lXxhQ1GAMKWqw2NhYVFVVWbRptVrEx8c7qSJqThhS1GCdO3fG8OHD4eLyz92psrISUVFRTqyKmguGFNlFXFyc+d9arRajR49G586dnVgRNRcMKbKL6dOnWxxJPRlaRA3BkCK7aNu2LV577TVoNBq4uLhg6tSpzi6JmgmGFNmNXq+HiGDChAlo06aNs8uhZoJ3QbBSUVERjh075uwylFZRUYFWrVqhe/fuyMjIcHY5SuvatSuGDRvm7DKaBL6Z00oZGRmIiYlxdhnUTERFRSEzM9PZZTQJPJKyETP96e7fvw9fX99at2k0GqSnpyM6OtrBVall+vTpzi6hSeGaFNlVXQFFVF8MKSJSGkOKiJTGkCIipTGkiEhpDCkiUhpDioiUxpAiIqUxpIhIaQwpIlIaQ4qIlMaQIiKlMaSISGkMKSJSGkOKiJTGkCIipfGmdw505coVbNu2DWlpabh8+bKzy7HJo0eP8Ic//AGnTp1Cz549MXPmzEa9d9TRo0dx6dIlizadTgdfX1/4+fmhX79+aN26daO9PqmDR1IOVFhYiOzsbBQVFTm7FJvcuXMH48aNw+3bt7Fs2TJcunQJL774Iq5cudJorzl8+HAEBAQgPj4eixcvRn5+PsrKynD69Gl88MEH8Pf3R0REBC5cuNBoNZAihKyvYbwpAAARIklEQVSSnp4u9piu5cuXi1artUNFjpOQkCCjR482P66srJT+/fvL5MmTbRoHgKSnp9v0HD8/PwkJCanR/pe//EU6duwo7u7ucuLECZvGdLaoqCiJiopydhlNBo+kHMzV1dXZJdjk/v37+O///m+MGzfO3Obi4oLExER89dVXOHPmTKO+vpubW63tYWFh+Pzzz1FWVobIyEiUl5c3ah3kPFyTamQmkwl79uzB6dOnMXr0aFRVVdXo8+DBA6SnpyM3Nxfdu3fH7Nmz4eXlBQC4ePEiUlNT8d5776GgoAAZGRlo3749Zs+ebRF4R44cQVZWFrp27QoXFxfMnTvXqvGf5YcffkBlZSU8PDws2nv37g0AOH36NAYMGGDzvNhDREQEwsLCcOjQIWRmZkKv1wNQez6pHpx9KNdU1Od07969exIWFiZr1qyRO3fuyI4dO8TNzc3idO+HH36Q119/Xb7++mv57rvvpG/fvtKjRw+5e/eupKamSocOHQSAfPXVVzJt2jSZOHGiAJBVq1aZx0hOTpa0tDQxGAyyc+dO8fLysmp8a1y4cEEAyIIFCyza//a3vwkAWbt2rdXzgXqc7nXs2LHW071q77zzjgCQOXPmiIj68ynC0z1bMaSsVJ+QWrBggUyZMsWibdKkSRYhFR4eLnv27DE/zsrKsvihSU5OFgCyb98+c58xY8ZIcHCwiIgYjUbx9/eXvLw88/akpCSrx38Wk8kkzz33nLRp08biB7F6PrZs2WLVOCKNE1JffPGFAJDw8HARUX8+RRhStuLpXiO5desWUlJS8PHHH1u09+/fH1lZWQCA69ev4+DBgxg4cCBOnjwJACgpKcGQIUNQWloKAPD09ATw+NSmWt++fc39XV1d4e3tjbFjx2Lr1q2YMGECVq5cafX4z6LT6ZCSkoLIyEiEhobizTffxMOHD80fbNmnT596zY+9GAwGAEBAQECTmE+yHUOqkZw5cwYmkwkdO3a0aNdoNOZ/5+fnAwCSk5PRrl27Wsdxcal5bcPT0xMVFRXmx5988gni4uIQERGBYcOGITU1FQEBAVaNb43XXnsNZ8+eRVZWFlq1aoVp06Zh69at6NWrF15++eV6j2sPeXl5AIDQ0NAmM59kG17dayQPHz4E8Pi3b12qr1ydOnWqzudbY+LEibh48SKWLFmCnJwcDBkyBLm5uXYbHwACAwMxf/58zJkzB9nZ2bh58ybWr19vEbqOZjQasX//fuh0OkydOrVJzSdZjyHVSJ5//nkAMJ/aPan6Cl9ISAi0Wi1Wr14No9Fo3l5cXIy0tDSrXsdgMCAlJQV+fn7YuHEjsrOzUVJSgp07d9pl/J+7f/8+PvzwQ3z66adO/7j0DRs2mMMkNDS0Sc4nWcHZi2JNRX0WzsePHy8AZPv27SIiUl5ebr6adPXqVTGZTLJw4UIBIEOHDpW0tDTZvn27hIeHy71790REZOnSpQJAHjx4YB532bJlotFopKysTB48eCCdOnWSR48embcHBwfLpk2bRESeOb4tbt26JYMGDZINGzbY/FwR2xfOjUajaLXaGgvnZWVlsmTJEnF1dZUVK1ZIWVmZeVtTmE8unNuGIWWl+oTUjRs3ZOTIkQJAgoODZfLkyaLX68XLy0sWLVokRUVFYjAYJD4+XgAIAPHx8TFfPdq7d68EBgYKAElKSpLCwkLZtWuXBAUFCQBZvny5FBQUiIeHh/Tr1082bdoka9askcTERDEajSIiTx3fWidPnpQPP/xQJk2aJL/73e9seu6TbAmpw4cPS2RkpAAQnU4ngwYNkqlTp0pkZKRMmjRJ5s2bJzk5OTWe1xTmkyFlG42IiKOO2pqyjIwMxMTEoD7TVVBQgMrKSvTq1QuXL1+Gv78/fHx8LPrcvn0bV69eRe/evWu8cfJpRASPHj1CZWUl8vPzERwcXOsbCxsy/tGjRxEYGIhf/OIXDVqD0mg0SE9Pd8hpoqrzCQDTp08HAPMVUno6Xt1zgB49epj/HRQUVGufdu3a1euKkUajMd8NYPDgwXX2a8j4I0aMsPl5zqbqfJLtuHBORErjkVQLde3aNSQmJj6zX0JCAuLi4hxQEVHtGFItVJcuXXDgwIFn9tPpuIuQc3EPbKE0Gg1atWrl7DKInolrUkSkNIYUESmNIUVESmNIEZHSGFJEpDSGFBEpjSFFREpjSBGR0hhSRKQ0hhQRKY1/FmOjjIwMZ5fQpB0/ftzZJThdUVERunTp4uwymgze9M5K1Te9I7KHqKgo3vTOSgwpsitH3n2TWgauSRGR0hhSRKQ0hhQRKY0hRURKY0gRkdIYUkSkNIYUESmNIUVESmNIEZHSGFJEpDSGFBEpjSFFREpjSBGR0hhSRKQ0hhQRKY0hRURKY0gRkdIYUkSkNIYUESmNIUVESmNIEZHSGFJEpDSGFBEpjSFFREpjSBGR0hhSRKQ0hhQRKY0hRURKY0gRkdIYUkSkNIYUESmNIUVEStM5uwBquo4cOYKysrIa7WfPnoWfn59F2wsvvABfX19HlUbNiEZExNlFUNMUHR2NzMzMZ/bz8PBAcXExPD09HVAVNTc83aN6i42NfWYfnU6HN954gwFF9caQonqbOHEivL29n9qnsrISer3eQRVRc8SQonpzc3NDVFQU3Nzc6uzj7e2N8PBwB1ZFzQ1Dihpk5syZMBqNtW5zdXXFjBkznhpiRM/ChXNqkKqqKnTo0AG3b9+udXt2djZGjRrl4KqoOeGRFDWIi4sLZs2aVevRUkBAAEaOHOmEqqg5YUhRg8XGxtY45XNzc0NCQgJcXLiLUcPwdI/sIjAwEFeuXLFoy8nJweDBg51UETUX/DVHdhEXFwdXV1fz427dujGgyC4YUmQXs2bNgslkAvD4VC8xMdHJFVFzwdM9sps+ffrg/PnzAIALFy4gJCTEyRVRc8AjKbKb+Ph4AED//v0ZUGQ3DCmymxkzZkCj0SAuLs7ZpVAzwtO9OoSGhiI3N9fZZVALcu7cOfTp08fZZSiH95N6iqioKEyfPt3ZZTQply5dQlBQkFV9MzMzcfz4cXz00UeNXJXaioqK8Oabbzq7DGUxpJ6iT58+iI6OdnYZzdb58+fxj3/8o8XP8T/+8Q+G1FNwTYqIlMaQIiKlMaSISGkMKSJSGkOKiJTGkCIipTGkiEhpDCkiUhpDioiUxpAiIqUxpIhIaQwpIlIaQ4qIlMaQIiKlMaSISGm8n1QjuXLlCrZt24a0tDRcvnzZ2eXUyzfffAOTyYSJEyfWuv306dP4/e9/j+eeew4zZ86El5dXo9Vy9OhRXLp0yaJNp9PB19cXfn5+6NevH1q3bt1or0/OwyOpRlJYWIjs7GwUFRU5uxSb/eUvf8G4ceMwbtw4/O///m+tfbZv346VK1fiX//1X+Hu7o7Ro0fj9u3bjVbT8OHDERAQgPj4eCxevBj5+fkoKyvD6dOn8cEHH8Df3x8RERG4cOFCo9VAzsEjqUYyZswYvPzyyzh58qSzS7HZiBEj0LNnzzpvA3z+/HkkJSUhLy8PnTp1Qnx8PNLS0vDuu+/i008/bZSaNBoNxo0bh7Zt2yIgIABr16612H7o0CHo9XoMGjQI2dnZeOmllxqlDnI8Hkk1oic/0bcpcXd3xy9+8Ys6ty9btgy9evVCp06dzG2vvvoqPv/8c1y7dq1Ra3Nzc6u1PSwsDJ9//jnKysoQGRmJ8vLyRq2DHIdHUnZkMpmwZ88enD59GqNHj0ZVVVWNPg8ePEB6ejpyc3PRvXt3zJ4927yWc/HiRaSmpuK9995DQUEBMjIy0L59e8yePdsi8I4cOYKsrCx07doVLi4umDt3rlXj20Kr1da57dSpUxgzZoxFW2BgIIxGIw4ePIg5c+bY/Hr2EBERgbCwMBw6dAiZmZnQ6/UAms6cU+14JGUn9+/fx4QJE5Cbm4vly5fj5s2bNT4FJT8/H3q9Ht26dUNCQgK2bt2KgQMH4t69e9ixYwdGjBiB999/HwcOHMDbb7+N48ePY+7cuVi3bp15jLfeegtXr17FypUr0aZNG4sb+D9tfFtpNBqL/1a7ffs2bt68CX9/f4v2wMBAAKixuO1ow4YNAwD8z//8D4CmNedUB6Fa9e7dW1avXm11/wULFsiUKVMs2iZNmiRardb8ODw8XPbs2WN+nJWVJQBk1apVIiKSnJwsAGTfvn3mPmPGjJHg4GARETEajeLv7y95eXnm7UlJSVaPb4uqqioBIGvXrrVo/+tf/yoA5N1337VoLygoEACSkJBg9WusXr1aevfubVNdHTt2lJCQkDq3f/HFFwJAwsPDRaRpzPm5c+cEgJw7d87q57QkPN2zg1u3biElJQUff/yxRXv//v2RlZUFALh+/ToOHjyIgQMHmhfTS0pKMGTIEJSWlgIAPD09ATw+banWt29fc39XV1d4e3tj7Nix2Lp1KyZMmICVK1daPb49yP//LNmfr7c9evQIANCxY0e7vVZ9GAwGAEBAQECzmfOWjiFlB2fOnIHJZKrxA/rkqVJ+fj4AIDk5Ge3atat1HBeXmmffnp6eqKioMD/+5JNPEBcXh4iICAwbNgypqakICAiwanx76NKlCwDg7t27Fu3V4dC3b99Ge21r5OXlAXj8CdTNZc5bOq5J2cHDhw8BPP7NWpfqq1KnTp2q8/nWmDhxIi5evIglS5YgJycHQ4YMQW5urt3Gf5bAwED4+fnV+F6vXLkCAE79mHCj0Yj9+/dDp9Nh6tSpzWbOWzqGlB08//zzAGA+tXtS9RW+kJAQaLVarF69Gkaj0by9uLgYaWlpVr2OwWBASkoK/Pz8sHHjRmRnZ6OkpAQ7d+60y/hPqj6tq/5vNTc3N8ycORN///vfLdq///57BAQEIDQ01ObXspcNGzaYwyQ0NLTJzTnVjiFlB6GhoRg/fjz279+P1NRUAI9/q3/33XcQEVy7dg3e3t6YN28eTpw4gVGjRuHLL79EamoqZs2ahdjYWADATz/9BOCf6zsAUFFRAZPJhPLyclRVVWH16tUoKysD8PhKVq9evRAQEIC2bds+c3xbVP/QVZ/GPemtt95CRUWFOahKSkqwbds2rF+/Hq1atbL5taxlMplQXFxco728vBxLly7F2rVrsWLFCqxfvx4ArJoTleac6uDcdXt12Xp178aNGzJy5EgBIMHBwTJ58mTR6/Xi5eUlixYtkqKiIjEYDBIfHy8ABID4+PiYrwzt3btXAgMDBYAkJSVJYWGh7Nq1S4KCggSALF++XAoKCsTDw0P69esnmzZtkjVr1khiYqIYjUYRkaeOb4tjx47JggULBID07NlTNm/eLCaTyaLPyZMnJSwsTP793/9dZs6cKf/xH/9h8+vYcnXv8OHDEhkZKQBEp9PJoEGDZOrUqRIZGSmTJk2SefPmSU5OTo3nNYU559W9p2NI1cHWkKp28eJFycvLk6qqKiksLJT79+/X6FNcXCw5OTlSWlpq09hVVVViMBjkwYMHkpOTIw8fPqy1X33Hr4/CwkKprKys13Pr8xaE+lJ5zhlST8ere3bWo0cP87/r+tu3du3a1etqkEajMf+l/+DBg+vsV9/x66Ou71E1zWnOWxquSRGR0ngk1QJcu3YNiYmJz+yXkJCAuLg4B1REZD2GVAvQpUsXHDhw4Jn9dDruDqQe7pUtgEajadS3BhA1Jq5JEZHSGFJEpDSGFBEpjSFFREpjSBGR0hhSRKQ0hhQRKY0hRURKY0gRkdIYUkSkNIYUESlNI/Kzm1gTgMe3BM7NzXV2GdSCnDt3zqkfZKEq/oFxHTZu3Ij79+87uwxqQbp27ersEpTEIykiUhrXpIhIaQwpIlIaQ4qIlKYDkOnsIoiI6vL/AAXfPQQpbTkFAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras.utils import plot_model\n",
    "\n",
    "plot_model(model, to_file='model.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "861e14df",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "36/36 [==============================] - 1s 19ms/step - loss: 1.0647 - accuracy: 0.3952 - val_loss: 0.9857 - val_accuracy: 0.5241\n",
      "Epoch 2/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.7104 - accuracy: 0.6334 - val_loss: 0.5840 - val_accuracy: 0.6898\n",
      "Epoch 3/50\n",
      "36/36 [==============================] - 0s 10ms/step - loss: 0.5388 - accuracy: 0.7431 - val_loss: 0.4845 - val_accuracy: 0.7781\n",
      "Epoch 4/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.4915 - accuracy: 0.7814 - val_loss: 0.4701 - val_accuracy: 0.7807\n",
      "Epoch 5/50\n",
      "36/36 [==============================] - 0s 10ms/step - loss: 0.5206 - accuracy: 0.7636 - val_loss: 0.4682 - val_accuracy: 0.8021\n",
      "Epoch 6/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.4338 - accuracy: 0.8046 - val_loss: 0.4117 - val_accuracy: 0.8209\n",
      "Epoch 7/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.3863 - accuracy: 0.8225 - val_loss: 0.3752 - val_accuracy: 0.8449\n",
      "Epoch 8/50\n",
      "36/36 [==============================] - 0s 10ms/step - loss: 0.3547 - accuracy: 0.8608 - val_loss: 0.3618 - val_accuracy: 0.8422\n",
      "Epoch 9/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.3203 - accuracy: 0.8849 - val_loss: 0.3410 - val_accuracy: 0.8636\n",
      "Epoch 10/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.3348 - accuracy: 0.8680 - val_loss: 0.3387 - val_accuracy: 0.8476\n",
      "Epoch 11/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.2647 - accuracy: 0.8974 - val_loss: 0.3208 - val_accuracy: 0.8663\n",
      "Epoch 12/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.2428 - accuracy: 0.9028 - val_loss: 0.3247 - val_accuracy: 0.8503\n",
      "Epoch 13/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.2377 - accuracy: 0.9063 - val_loss: 0.2706 - val_accuracy: 0.8797\n",
      "Epoch 14/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.2073 - accuracy: 0.9153 - val_loss: 0.3120 - val_accuracy: 0.8824\n",
      "Epoch 15/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.2042 - accuracy: 0.9153 - val_loss: 0.2731 - val_accuracy: 0.8930\n",
      "Epoch 16/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.1708 - accuracy: 0.9340 - val_loss: 0.3448 - val_accuracy: 0.8636\n",
      "Epoch 17/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.1590 - accuracy: 0.9402 - val_loss: 0.2354 - val_accuracy: 0.9198\n",
      "Epoch 18/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.1255 - accuracy: 0.9492 - val_loss: 0.2180 - val_accuracy: 0.9251\n",
      "Epoch 19/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.1272 - accuracy: 0.9527 - val_loss: 0.2567 - val_accuracy: 0.9144\n",
      "Epoch 20/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.1191 - accuracy: 0.9581 - val_loss: 0.2422 - val_accuracy: 0.9118\n",
      "Epoch 21/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0931 - accuracy: 0.9652 - val_loss: 0.2967 - val_accuracy: 0.9064\n",
      "Epoch 22/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0837 - accuracy: 0.9723 - val_loss: 0.3757 - val_accuracy: 0.8690\n",
      "Epoch 23/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0869 - accuracy: 0.9706 - val_loss: 0.2744 - val_accuracy: 0.9118\n",
      "Epoch 24/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0712 - accuracy: 0.9732 - val_loss: 0.2584 - val_accuracy: 0.9144\n",
      "Epoch 25/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0650 - accuracy: 0.9804 - val_loss: 0.3000 - val_accuracy: 0.9011\n",
      "Epoch 26/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0524 - accuracy: 0.9848 - val_loss: 0.2689 - val_accuracy: 0.9091\n",
      "Epoch 27/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0841 - accuracy: 0.9634 - val_loss: 0.3085 - val_accuracy: 0.9037\n",
      "Epoch 28/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0673 - accuracy: 0.9732 - val_loss: 0.2785 - val_accuracy: 0.9144\n",
      "Epoch 29/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0347 - accuracy: 0.9911 - val_loss: 0.2498 - val_accuracy: 0.9118\n",
      "Epoch 30/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0293 - accuracy: 0.9946 - val_loss: 0.3549 - val_accuracy: 0.8930\n",
      "Epoch 31/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0268 - accuracy: 0.9938 - val_loss: 0.3179 - val_accuracy: 0.9144\n",
      "Epoch 32/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0292 - accuracy: 0.9929 - val_loss: 0.2627 - val_accuracy: 0.9144\n",
      "Epoch 33/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0321 - accuracy: 0.9893 - val_loss: 0.2908 - val_accuracy: 0.9171\n",
      "Epoch 34/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0151 - accuracy: 0.9973 - val_loss: 0.2891 - val_accuracy: 0.9225\n",
      "Epoch 35/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0102 - accuracy: 1.0000 - val_loss: 0.2639 - val_accuracy: 0.9118\n",
      "Epoch 36/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0079 - accuracy: 1.0000 - val_loss: 0.2923 - val_accuracy: 0.9144\n",
      "Epoch 37/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0058 - accuracy: 1.0000 - val_loss: 0.3063 - val_accuracy: 0.9118\n",
      "Epoch 38/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0050 - accuracy: 1.0000 - val_loss: 0.3088 - val_accuracy: 0.9091\n",
      "Epoch 39/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0044 - accuracy: 1.0000 - val_loss: 0.3123 - val_accuracy: 0.9144\n",
      "Epoch 40/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0043 - accuracy: 1.0000 - val_loss: 0.3246 - val_accuracy: 0.9171\n",
      "Epoch 41/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0033 - accuracy: 1.0000 - val_loss: 0.3228 - val_accuracy: 0.9144\n",
      "Epoch 42/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0029 - accuracy: 1.0000 - val_loss: 0.3201 - val_accuracy: 0.9118\n",
      "Epoch 43/50\n",
      "36/36 [==============================] - 0s 10ms/step - loss: 0.0026 - accuracy: 1.0000 - val_loss: 0.3225 - val_accuracy: 0.9144\n",
      "Epoch 44/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0023 - accuracy: 1.0000 - val_loss: 0.3124 - val_accuracy: 0.9118\n",
      "Epoch 45/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0021 - accuracy: 1.0000 - val_loss: 0.3206 - val_accuracy: 0.9118\n",
      "Epoch 46/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0020 - accuracy: 1.0000 - val_loss: 0.3232 - val_accuracy: 0.9091\n",
      "Epoch 47/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0019 - accuracy: 1.0000 - val_loss: 0.3300 - val_accuracy: 0.9198\n",
      "Epoch 48/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0015 - accuracy: 1.0000 - val_loss: 0.3325 - val_accuracy: 0.9144\n",
      "Epoch 49/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0014 - accuracy: 1.0000 - val_loss: 0.3248 - val_accuracy: 0.9144\n",
      "Epoch 50/50\n",
      "36/36 [==============================] - 0s 9ms/step - loss: 0.0013 - accuracy: 1.0000 - val_loss: 0.3340 - val_accuracy: 0.9171\n"
     ]
    }
   ],
   "source": [
    "# Compile the model\n",
    "model.compile(optimizer='Adam',\n",
    "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "# Train the model\n",
    "history = model.fit(X_train, y_train, epochs=50,\n",
    "                    validation_data=(X_test, y_test))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "c02cabcb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "pd.DataFrame(history.history).plot(figsize=(5, 3))\n",
    "plt.grid(True)\n",
    "plt.gca().set_ylim(0, 1.1)\n",
    "plt.xlabel('epochs')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "26c6aac4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "36/36 [==============================] - 1s 15ms/step - loss: 1.0856 - accuracy: 0.3515 - val_loss: 1.0589 - val_accuracy: 0.5000\n",
      "Epoch 2/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.9743 - accuracy: 0.5602 - val_loss: 0.8589 - val_accuracy: 0.6524\n",
      "Epoch 3/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.6763 - accuracy: 0.7208 - val_loss: 0.6305 - val_accuracy: 0.6925\n",
      "Epoch 4/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.5431 - accuracy: 0.7645 - val_loss: 0.5758 - val_accuracy: 0.7273\n",
      "Epoch 5/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.5778 - accuracy: 0.7377 - val_loss: 0.5298 - val_accuracy: 0.7781\n",
      "Epoch 6/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.4665 - accuracy: 0.7797 - val_loss: 0.4687 - val_accuracy: 0.8021\n",
      "Epoch 7/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.4327 - accuracy: 0.8046 - val_loss: 0.4464 - val_accuracy: 0.7888\n",
      "Epoch 8/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.4104 - accuracy: 0.8198 - val_loss: 0.4903 - val_accuracy: 0.7754\n",
      "Epoch 9/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.4084 - accuracy: 0.8136 - val_loss: 0.4181 - val_accuracy: 0.8182\n",
      "Epoch 10/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.3796 - accuracy: 0.8332 - val_loss: 0.4195 - val_accuracy: 0.8235\n",
      "Epoch 11/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.3582 - accuracy: 0.8439 - val_loss: 0.4388 - val_accuracy: 0.8155\n",
      "Epoch 12/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.3492 - accuracy: 0.8421 - val_loss: 0.5015 - val_accuracy: 0.7701\n",
      "Epoch 13/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.3661 - accuracy: 0.8368 - val_loss: 0.3854 - val_accuracy: 0.8316\n",
      "Epoch 14/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.3720 - accuracy: 0.8323 - val_loss: 0.4418 - val_accuracy: 0.8155\n",
      "Epoch 15/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.3420 - accuracy: 0.8582 - val_loss: 0.4331 - val_accuracy: 0.8128\n",
      "Epoch 16/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.3233 - accuracy: 0.8635 - val_loss: 0.3654 - val_accuracy: 0.8449\n",
      "Epoch 17/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.3118 - accuracy: 0.8715 - val_loss: 0.3527 - val_accuracy: 0.8556\n",
      "Epoch 18/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.3077 - accuracy: 0.8715 - val_loss: 0.3704 - val_accuracy: 0.8503\n",
      "Epoch 19/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2911 - accuracy: 0.8760 - val_loss: 0.4082 - val_accuracy: 0.8316\n",
      "Epoch 20/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2871 - accuracy: 0.8796 - val_loss: 0.5876 - val_accuracy: 0.7513\n",
      "Epoch 21/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.4743 - accuracy: 0.8109 - val_loss: 0.3774 - val_accuracy: 0.8583\n",
      "Epoch 22/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.3094 - accuracy: 0.8796 - val_loss: 0.3418 - val_accuracy: 0.8717\n",
      "Epoch 23/50\n",
      "36/36 [==============================] - 0s 6ms/step - loss: 0.2786 - accuracy: 0.8983 - val_loss: 0.3627 - val_accuracy: 0.8422\n",
      "Epoch 24/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2670 - accuracy: 0.9010 - val_loss: 0.3207 - val_accuracy: 0.8743\n",
      "Epoch 25/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2581 - accuracy: 0.8956 - val_loss: 0.3297 - val_accuracy: 0.8663\n",
      "Epoch 26/50\n",
      "36/36 [==============================] - 0s 6ms/step - loss: 0.2626 - accuracy: 0.8992 - val_loss: 0.3545 - val_accuracy: 0.8422\n",
      "Epoch 27/50\n",
      "36/36 [==============================] - 0s 6ms/step - loss: 0.2407 - accuracy: 0.9117 - val_loss: 0.3218 - val_accuracy: 0.8636\n",
      "Epoch 28/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2539 - accuracy: 0.9028 - val_loss: 0.3081 - val_accuracy: 0.8743\n",
      "Epoch 29/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2406 - accuracy: 0.9081 - val_loss: 0.3035 - val_accuracy: 0.8770\n",
      "Epoch 30/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2279 - accuracy: 0.9179 - val_loss: 0.3168 - val_accuracy: 0.8743\n",
      "Epoch 31/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2228 - accuracy: 0.9161 - val_loss: 0.4096 - val_accuracy: 0.8235\n",
      "Epoch 32/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2804 - accuracy: 0.8858 - val_loss: 0.3789 - val_accuracy: 0.8476\n",
      "Epoch 33/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2288 - accuracy: 0.9063 - val_loss: 0.3153 - val_accuracy: 0.8770\n",
      "Epoch 34/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2248 - accuracy: 0.9108 - val_loss: 0.3575 - val_accuracy: 0.8690\n",
      "Epoch 35/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2216 - accuracy: 0.9117 - val_loss: 0.3216 - val_accuracy: 0.8770\n",
      "Epoch 36/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2181 - accuracy: 0.9188 - val_loss: 0.3320 - val_accuracy: 0.8770\n",
      "Epoch 37/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1967 - accuracy: 0.9251 - val_loss: 0.2936 - val_accuracy: 0.8850\n",
      "Epoch 38/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1892 - accuracy: 0.9322 - val_loss: 0.3038 - val_accuracy: 0.8877\n",
      "Epoch 39/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2128 - accuracy: 0.9161 - val_loss: 0.3695 - val_accuracy: 0.8396\n",
      "Epoch 40/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1830 - accuracy: 0.9358 - val_loss: 0.3133 - val_accuracy: 0.8797\n",
      "Epoch 41/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1773 - accuracy: 0.9295 - val_loss: 0.3684 - val_accuracy: 0.8583\n",
      "Epoch 42/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1667 - accuracy: 0.9384 - val_loss: 0.2998 - val_accuracy: 0.8930\n",
      "Epoch 43/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1683 - accuracy: 0.9420 - val_loss: 0.3533 - val_accuracy: 0.8717\n",
      "Epoch 44/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1966 - accuracy: 0.9304 - val_loss: 0.2939 - val_accuracy: 0.8743\n",
      "Epoch 45/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1533 - accuracy: 0.9402 - val_loss: 0.2904 - val_accuracy: 0.8957\n",
      "Epoch 46/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1532 - accuracy: 0.9447 - val_loss: 0.3798 - val_accuracy: 0.8476\n",
      "Epoch 47/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.2522 - accuracy: 0.8983 - val_loss: 0.3399 - val_accuracy: 0.8529\n",
      "Epoch 48/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1620 - accuracy: 0.9465 - val_loss: 0.3673 - val_accuracy: 0.8717\n",
      "Epoch 49/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1469 - accuracy: 0.9465 - val_loss: 0.2892 - val_accuracy: 0.8850\n",
      "Epoch 50/50\n",
      "36/36 [==============================] - 0s 7ms/step - loss: 0.1341 - accuracy: 0.9509 - val_loss: 0.3399 - val_accuracy: 0.8850\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# smaller network\n",
    "\n",
    "model = Sequential()\n",
    "\n",
    "model.add(Conv2D(5, (3, 3), (2, 2), activation='relu', input_shape=(64, 64, 3)))\n",
    "model.add(MaxPooling2D((3, 3), strides=(2, 2)))\n",
    "model.add(Conv2D(10, (5, 5), activation='relu', padding=\"same\"))\n",
    "model.add(MaxPooling2D((3, 3), strides=(2, 2)))\n",
    "model.add(Flatten())\n",
    "model.add(Dense(16, activation='relu'))\n",
    "model.add(Dense(3))\n",
    "\n",
    "model.compile(optimizer='Adam',\n",
    "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "\n",
    "history = model.fit(X_train, y_train, epochs=50,\n",
    "                    validation_data=(X_test, y_test))\n",
    "\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "pd.DataFrame(history.history).plot(figsize=(5, 3))\n",
    "plt.grid(True)\n",
    "plt.gca().set_ylim(0, 1.1)\n",
    "plt.xlabel('epochs')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b0331a59",
   "metadata": {},
   "source": [
    "### 1D Convolutional Neural Network \n",
    "* A 1D Convolutional Neural Network (CNN) is a type of neural network architecture designed to analyze and extract patterns from one-dimensional sequences of data, such as time series or sequential data\n",
    "\n",
    "    * 1D CNNs apply convolutional filters (kernels) to the input sequence to capture local patterns and features\n",
    "    * These filters slide along the sequence, similar to how 2D CNNs slide filters over images\n",
    "\n",
    "\n",
    "* Next, we will investigate 1D CNNs\n",
    "\n",
    "    * We generate synthetic sensor data to simulate sensor readings or measurements\n",
    "\n",
    "    * The synthetic sensor data consists of two classes (or patterns), referred to as \"class_0_data\" and \"class_1_data\"\n",
    "    \n",
    "    * Each class contains data samples with a length of 100 data points (or time steps)\n",
    "\n",
    "    * The data for \"class_0\" is generated using random values from a standard normal distribution (mean=0, standard deviation=1)\n",
    "\n",
    "    * The data for \"class_1\" is generated similarly but with a different mean and standard deviation to create a distinct pattern"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "18944b6f",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras import layers, models\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Generate synthetic sensor data\n",
    "np.random.seed(0)\n",
    "num_samples = 1000\n",
    "signal_length = 100\n",
    "num_classes = 2\n",
    "\n",
    "# Create a synthetic dataset with two classes (patterns)\n",
    "class_0_data = np.random.randn(num_samples // 2, signal_length)\n",
    "class_1_data = 2 * np.random.randn(num_samples // 2, signal_length) \n",
    "\n",
    "# Create labels for the dataset\n",
    "class_0_labels = np.zeros((num_samples // 2,), dtype=np.int32)\n",
    "class_1_labels = np.ones((num_samples // 2,), dtype=np.int32)\n",
    "labels = np.concatenate([class_0_labels, class_1_labels])\n",
    "\n",
    "# Combine the data from both classes\n",
    "data = np.concatenate([class_0_data, class_1_data])\n",
    "\n",
    "# Shuffle the data and labels\n",
    "shuffle_indices = np.arange(num_samples)\n",
    "np.random.shuffle(shuffle_indices)\n",
    "data = data[shuffle_indices]\n",
    "labels = labels[shuffle_indices]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "c693b803",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(data[6], 'r-' , label = labels[6])  \n",
    "plt.plot(data[1], 'b', label= labels[1]) \n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "4a38f2d9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "18/18 [==============================] - 1s 26ms/step - loss: 0.6706 - accuracy: 0.5536 - val_loss: 0.6779 - val_accuracy: 0.4786\n",
      "Epoch 2/10\n",
      "18/18 [==============================] - 0s 8ms/step - loss: 0.5747 - accuracy: 0.6232 - val_loss: 0.7344 - val_accuracy: 0.5429\n",
      "Epoch 3/10\n",
      "18/18 [==============================] - 0s 7ms/step - loss: 0.5112 - accuracy: 0.7446 - val_loss: 0.6081 - val_accuracy: 0.5714\n",
      "Epoch 4/10\n",
      "18/18 [==============================] - 0s 7ms/step - loss: 0.4012 - accuracy: 0.8357 - val_loss: 0.5372 - val_accuracy: 0.7429\n",
      "Epoch 5/10\n",
      "18/18 [==============================] - 0s 7ms/step - loss: 0.2648 - accuracy: 0.9482 - val_loss: 0.3157 - val_accuracy: 0.9143\n",
      "Epoch 6/10\n",
      "18/18 [==============================] - 0s 7ms/step - loss: 0.1340 - accuracy: 0.9911 - val_loss: 0.1861 - val_accuracy: 0.9500\n",
      "Epoch 7/10\n",
      "18/18 [==============================] - 0s 6ms/step - loss: 0.0547 - accuracy: 1.0000 - val_loss: 0.0930 - val_accuracy: 0.9857\n",
      "Epoch 8/10\n",
      "18/18 [==============================] - 0s 6ms/step - loss: 0.0283 - accuracy: 1.0000 - val_loss: 0.0566 - val_accuracy: 0.9929\n",
      "Epoch 9/10\n",
      "18/18 [==============================] - 0s 8ms/step - loss: 0.0163 - accuracy: 1.0000 - val_loss: 0.0445 - val_accuracy: 0.9929\n",
      "Epoch 10/10\n",
      "18/18 [==============================] - 0s 6ms/step - loss: 0.0117 - accuracy: 1.0000 - val_loss: 0.0402 - val_accuracy: 0.9929\n",
      "10/10 - 0s - loss: 0.0228 - accuracy: 1.0000 - 39ms/epoch - 4ms/step\n",
      "\n",
      "Test accuracy: 1.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 400x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Split the dataset into training and testing sets\n",
    "train_split = 0.7\n",
    "split_index = int(train_split * num_samples)\n",
    "train_data, test_data = data[:split_index], data[split_index:]\n",
    "train_labels, test_labels = labels[:split_index], labels[split_index:]\n",
    "\n",
    "# Create a 1D CNN model\n",
    "model = models.Sequential()\n",
    "model.add(layers.Conv1D(32, 3, activation='relu', input_shape=(signal_length, 1)))\n",
    "model.add(layers.MaxPooling1D(2))\n",
    "model.add(layers.Flatten())\n",
    "model.add(layers.Dense(32, activation='relu'))\n",
    "model.add(layers.Dense(num_classes, activation='softmax'))\n",
    "\n",
    "# Compile the model\n",
    "model.compile(optimizer='adam',\n",
    "              loss='sparse_categorical_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "# Reshape data to fit the model's input shape\n",
    "train_data = train_data[:, :, np.newaxis]\n",
    "test_data = test_data[:, :, np.newaxis]\n",
    "\n",
    "# Train the model\n",
    "history = model.fit(train_data, train_labels, epochs=10, validation_split=0.2)\n",
    "\n",
    "# Evaluate the model\n",
    "test_loss, test_acc = model.evaluate(test_data, test_labels, verbose=2)\n",
    "print(\"\\nTest accuracy:\", test_acc)\n",
    "\n",
    "# Plot training history\n",
    "plt.plot(history.history['accuracy'], label='Accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.ylim([0, 1.01])\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cf2e9c76",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "99f7ea5f",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}