From 210e284a4d3b0b726caf060349ec8b845b0662e1 Mon Sep 17 00:00:00 2001 From: Anjalijain123 <85122785+Anjalijain123@users.noreply.github.com> Date: Mon, 6 Jun 2022 11:33:24 +0530 Subject: [PATCH 1/4] Add files via upload --- Assignment 1/A1_200132.ipynb | 282 +++++++++++++++++++++++++++++++++++ 1 file changed, 282 insertions(+) create mode 100644 Assignment 1/A1_200132.ipynb diff --git a/Assignment 1/A1_200132.ipynb b/Assignment 1/A1_200132.ipynb new file mode 100644 index 0000000..de2aec8 --- /dev/null +++ b/Assignment 1/A1_200132.ipynb @@ -0,0 +1,282 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "name": "A1_200132", + "provenance": [], + "collapsed_sections": [] + }, + "kernelspec": { + "name": "python3", + "display_name": "Python 3" + }, + "language_info": { + "name": "python" + } + }, + "cells": [ + { + "cell_type": "markdown", + "source": [ + "### **Aim** \n", + "The motive of this assignment is to make predictions using **Linear Regression**. To make sure you truly understand how the underlying algorithm works, you are to implement it from scratch." + ], + "metadata": { + "id": "RB2d1J1f1CF7" + } + }, + { + "cell_type": "markdown", + "source": [ + "### Generating the dataset \n", + "Run the cell below to create the dataset. It further splits the available data into training and testing. Please do not edit this cell.\n" + ], + "metadata": { + "id": "a_S80lf6H4Xv" + } + }, + { + "cell_type": "code", + "source": [ + "from sklearn import datasets\n", + "from sklearn.model_selection import train_test_split\n", + "\n", + "# Generate the data\n", + "X, y = datasets.make_regression(n_samples=100, n_features=5, noise=20, random_state=4)\n", + "\n", + "# Split the data\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1234)\n" + ], + "metadata": { + "id": "yX0zqXcHIQHP" + }, + "execution_count": 1, + "outputs": [] + }, + { + "cell_type": "markdown", + "source": [ + "### Visualizing the data \n", + "Use `matplotlib` to visualize the given data." + ], + "metadata": { + "id": "Zj4rrRXGJBXy" + } + }, + { + "cell_type": "code", + "source": [ + "import matplotlib.pyplot as plt\n", + "for i in range(0,5):\n", + " plt.scatter(X[:,1],y)\n", + " plt.show\n", + "# Your code here" + ], + "metadata": { + "id": "zxfi8dkBJOUi", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 265 + }, + "outputId": "18f86128-4b6d-4622-bc8f-43f687e64645" + }, + "execution_count": 2, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "You should be able to see the linear relations between `y` and the features in vector `X`." + ], + "metadata": { + "id": "r7vndSBAJceF" + } + }, + { + "cell_type": "markdown", + "source": [ + "### Gradient Descent Review \n", + "1. #### Cost function\n", + "Define the `cost function` to measure the difference between predictions and target outputs. Here, we are working with first degree polynomial, so derivatives are easy to calculate. ( Linear function `y = wx +b` ) \n", + "\n", + "$$Error = \\frac{1}{N}\\sum_{i=1}^N (y_i - \\overline{y}_i)^2 = \\frac{1}{N}\\sum_{i=1}^N (y_i - (x_iw+b))^2 $$ \n", + "\n", + " where `N` is the number of samples \n", + " \n", + "\n", + "\n", + "2. #### Compute the derivative\n", + "$$\\frac{\\delta Error}{\\delta w} = \\frac{2}{N}\\sum_{i=1}^N -x_i(y_i -(m x_i +b )) $$\n", + "$$\\frac{\\delta Error}{\\delta b} = \\frac{2}{N}\\sum_{i=1}^N -(y_i -(m x_i +b )) $$\n", + "3.

Update current parameters

\n", + "$$ w:= w- learning\\_rate \\cdot \\frac{\\delta Error}{\\delta w} $$ \n", + "$$ b:= b- learning\\_rate \\cdot \\frac{\\delta Error}{\\delta b} $$ \n", + "4.

Repeat until it fits good enough

\n" + ], + "metadata": { + "id": "b4I9Z3epNvBM" + } + }, + { + "cell_type": "markdown", + "source": [ + "### Model definition\n", + "\n", + "Complete the functions in the class below. Hints provided at appropriate places." + ], + "metadata": { + "id": "kBtUcOVnJu-I" + } + }, + { + "cell_type": "code", + "source": [ + "import numpy as np\n", + "\n", + "class LinearRegression:\n", + "\n", + " # The __init__ is called when we make any object of our class. Here, you are to specify the default values for \n", + " # Learning Rate, Number of Iterations, Weights and Biases. It doesn't return anything.\n", + " # Hint: Google what a `self pointer` is and figure out how it can be used here.\n", + " def __init__(self, learning_rate=0.001, n_iters=1000):\n", + " self.learning_rate = learning_rate\n", + " self.n_iters = n_iters\n", + " self.weights = None\n", + " self.bias = None\n", + " \n", + "\n", + " # Uncomment this when you're done with this function\n", + "\n", + "\n", + " # The following function would be the heart of the model. This is where the training would happen. \n", + " # You're supposed to iterate and keep on updating the weights and biases according to the steps of Gradient Descent.\n", + " def fit(self, X, y):\n", + " n_samples, n_features = X.shape\n", + " # init parameters\n", + " self.weights = np.zeros(n_features)\n", + " self.bias = 0\n", + " # Gradient Descent code goes here\n", + " for _ in range(self.n_iters):\n", + " y_predicted = np.dot(X, self.weights) + self.bias\n", + " \n", + " dw = (1 / n_samples) * np.dot(X.T, (y_predicted - y))\n", + " db = (1 / n_samples) * np.sum(y_predicted - y)\n", + " self.weights -= self.learning_rate * dw\n", + " self.bias -= self.learning_rate * db\n", + "\n", + " # Uncomment this when you're done with this function\n", + " \n", + " \n", + " # This function will be called after our model has been trained and we are predicting on unseen data\n", + " # What is our prediction? Just return that\n", + " def predict(self, X):\n", + " y_fin = np.dot(X, self.weights) + self.bias\n", + " return y_fin\n", + "\n", + "\n", + " \n", + "\n" + ], + "metadata": { + "id": "dGnFNPJx3I28" + }, + "execution_count": 3, + "outputs": [] + }, + { + "cell_type": "markdown", + "source": [ + "### Initializing, Training & Predictions" + ], + "metadata": { + "id": "EvyInkTKPn7W" + } + }, + { + "cell_type": "code", + "source": [ + "# Now, we make an object of our custom class.\n", + "\n", + "regressor = LinearRegression(learning_rate=0.01, n_iters=1000)# You may pass the custom parameters or let the default values take it ahead\n", + "regressor.fit(X_train, y_train)\n", + "\n", + "\n", + "# Call the fit method on the object to train (pass appropriate part of dataset)\n", + "\n", + "\n", + "# Now, let's see our what our model predicts\n", + "predictions = regressor.predict(X_test) # pass appropriate part of dataset" + ], + "metadata": { + "id": "nvItUpAkHTiv" + }, + "execution_count": 4, + "outputs": [] + }, + { + "cell_type": "markdown", + "source": [ + "### Evaluate the model \n", + "\n", + "Return [Mean Squared Error](https://en.wikipedia.org/wiki/Mean_squared_error) & [R2 Score](https://www.ncl.ac.uk/webtemplate/ask-assets/external/maths-resources/statistics/regression-and-correlation/coefficient-of-determination-r-squared.html#:~:text=%C2%AFy) from the functions below." + ], + "metadata": { + "id": "tzK6cq8eRD4Q" + } + }, + { + "cell_type": "code", + "source": [ + "def mean_squared_error(y_true, y_pred):\n", + " # return the mean squared error\n", + " return np.mean((y_true - y_pred) ** 2) # Uncomment this when you're done with this function\n", + "\n", + "\n", + "def r2_score(y_true, y_pred):\n", + " # return the r2 score\n", + " corr_matrix = np.corrcoef(y_true, y_pred)\n", + " corr = corr_matrix[0, 1]\n", + " return corr ** 2 # Uncomment this when you're done with this function\n", + " \n", + "\n", + "mse = mean_squared_error(y_test, predictions) # Pass appropriate parts of dataset\n", + "print(\"MSE:\", mse)\n", + "\n", + "accu = r2_score(y_test, predictions) # Pass appropriate parts of dataset\n", + "print(\"Accuracy:\", accu)" + ], + "metadata": { + "id": "WqkrvDzcRF5m", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "4ca01759-263e-47a4-dbaa-6f9cd07f1326" + }, + "execution_count": 5, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "MSE: 390.6026570894989\n", + "Accuracy: 0.9637788781894968\n" + ] + } + ] + } + ] +} \ No newline at end of file From 5867accc228d32254237fb03ae711685ab61098e Mon Sep 17 00:00:00 2001 From: Anjalijain123 <85122785+Anjalijain123@users.noreply.github.com> Date: Thu, 16 Jun 2022 22:14:08 +0530 Subject: [PATCH 2/4] Add files via upload --- Assignment 2/A2_200132.ipynb | 652 +++++++++++++++++++++++++++++++++++ 1 file changed, 652 insertions(+) create mode 100644 Assignment 2/A2_200132.ipynb diff --git a/Assignment 2/A2_200132.ipynb b/Assignment 2/A2_200132.ipynb new file mode 100644 index 0000000..37acbe0 --- /dev/null +++ b/Assignment 2/A2_200132.ipynb @@ -0,0 +1,652 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "UxcaEbrCy1g_" + }, + "source": [ + "# Assigment 2: Deep Learning" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "h2JON-_Oy79w" + }, + "source": [ + "## Generate Dataset\n", + "\n", + "This is the same code from Assignment 1" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": { + "id": "hgpG3WDuypfa" + }, + "outputs": [], + "source": [ + "from sklearn import datasets\n", + "from sklearn.model_selection import train_test_split\n", + "\n", + "# Generate the data\n", + "X, y = datasets.make_regression(n_samples=100, n_features=5, noise=5, random_state=4)\n", + "\n", + "# Split the data\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1234)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "r6it-Rm7zD1Y" + }, + "source": [ + "## Visualize Dataset\n", + "This is the same code from Assignment 1" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "metadata": { + "id": "UautPVj1yzaQ", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "outputId": "18ee9207-94c4-463d-8656-bdacb3ab34cd" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAD4CAYAAADlwTGnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAaEUlEQVR4nO3db4xcV3kG8OfxZhM2ArFJs0qTtY0tZJw6BDCsQip/aRKoHf45hIISIQgQyaqUqKRChnWRSvhQZSVLSFBSWgsiQLIIAYfFrYMWB7tCTWvImnX+OLHBTRTsIZBFsEDjVbzevP0wd+zZ2Xtn5s69d8659zw/ycrundmdc2ey973nPe85h2YGEREJzwrXDRARETcUAEREAqUAICISKAUAEZFAKQCIiATqAtcN6MZll11ma9ascd0MEZFSOXz48G/NbCTp8VIEgDVr1mB6etp1M0RESoXk8+0eVwpIRCRQCgAiIoFSABARCZQCgIhIoBQAREQCVYoqIBGRXkzO1LBz6jh+NTePK4eHsH3zety8cdR1s7yhACAilTQ5U8OOh57E/MIiAKA2N48dDz0JAKUIAv0IXkoBiUgl7Zw6fu7i3zC/sIidU8cdtah7jeBVm5uH4Xzwmpyp5fo6CgAiUkm/mptPddwn/QpeCgAiUklXDg+lOu6TfgUvBQARqaTtm9djaHBgybGhwQFs37zeUYu616/gpQAgIpV088ZR3HvLNRgdHgIBjA4P4d5brinFAHC/gpeqgCRoKhOstps3jpby82y0uej/NxUAJFhlLxOUautH8FIKSIJV5jJBkTwoAEiwylwmKJIHBQAJVpnLBEXyoAAgwSpzmaBIHjIHAJKrSB4k+TTJoyQ/GR2/lOR+kr+I/ntJdJwkv0TyBMknSL41axtEelHmMkGRPORRBXQWwKfM7GckXwPgMMn9AD4G4EdmNkFyHMA4gM8AuAnAuujf2wF8JfqvSN+VtUxQihVKeXDmHoCZvWBmP4u+/hOAZwCMAtgK4BvR074B4Obo660Avml1hwAMk7wiaztERPLQr4XYfJDrGADJNQA2AvgJgMvN7IXooV8DuDz6ehTAyaYfOxUda/1d20hOk5yenZ3Ns5kiIolCKg/ObSIYyVcD2APgbjP7I8lzj5mZkbQ0v8/MdgHYBQBjY2OpflZEys1lCiak8uBcegAkB1G/+O82s4eiw79ppHai/74YHa8BWNX04yujYyIizlMwIZUH51EFRABfA/CMmX2h6aG9AG6Pvr4dwPebjn80qga6DsAfmlJFIlIRkzM1bJo4gLXj+7Bp4kDXF3DXKZiQyoPzSAFtAvARAE+SPBId+wcAEwAeJHkHgOcBfCh67GEA7wJwAsBpAB/PoQ0i4pG4dZa2f+dxfP7fj2Lu9ELbtI7rFEy/FmLzQeYAYGb/BYAJD98Y83wDcGfW1xURf8XdxS+8Yvj96QUA7Rfeu3J4CLWYi30/UzChlAdrJrCI5K6bu/WktE5IKRjXFABEJHfd3q3HBQrN0O4f7QcgIrnbvnn9kjGAJEmBIpQUjGsKACKSu9aB1NcODeKlM2exsHh+So/SOu4pAIhIIVrv4kNZX6dMFABEpC+U1vGPBoFFRAKlHoBICSh9IkVQABDxXNys2qRJVCJpKACIeK7d2jg+BgD1VspDAUDEc67XxmnV7gKv3kq5aBBYxHM+LU/caalm1yt5SjoKACKe82ltnE4X+DS9lV6Xi5b8KAUk4jmflifudIHvdiVPpYr8oAAg0me9DJL6Momq0wU+bg2guN5K2Qa2q0opIJE+cr3dYVad0lHdruTp28B2qNQDEOmjst/5dpOO6qa34sOmL6IAINJXVbjzzSMd1W2qSIqlACDSQZ4Tm3TnW+fTwHbIFABE2si7WqXMd755z/D1ZWA7ZBoEFmkj74lNZd3usOyD1xJPPQCJpfVc6orI2Rd551vU51b2wWuJpwAgy2iSznllytkX+blVYfBallMKSJbRei7n+bQMQydFfm4+rUck+VEAkGV0t3demXL2RX5uZQqE0j2lgGSZMqU9+qEs1SrtPresYwMq26wmBQBZpsyliiFL+tyuv2okl7GBsgRC6Z4CQIn0qzJHd3t+6vT5J31uquCRJDQz123oaGxszKanp103w6nWCg+gfnfnaz5a8pXl8187vg9xf+UE8NzEu/NtqHiF5GEzG0t6XIPAJaHKnLBl+fzzquDRBi7VowBQEqrMCVuWzz+PCh7NBK6mXAIAyftJvkjyqaZjl5LcT/IX0X8viY6T5JdIniD5BMm35tGGqlMddtiyfP55lLKqB1pNefUAvg5gS8uxcQA/MrN1AH4UfQ8ANwFYF/3bBuArObWh0lSHHbbrrxoBW46l+fxv3jiKR8dvwHMT78aj4zekHjdSD7SacqkCMrMfk1zTcngrgL+Kvv4GgP8E8Jno+DetPvp8iOQwySvM7IU82lJV/arM0RpA/pmcqWHP4dqSgVwC+MDb+leWqbkh1VRkGejlTRf1XwO4PPp6FMDJpuedio4tCQAkt6HeQ8Dq1asLbGZ5FF2HrTWA/BSXfjEAB4/N9q0NmhtSTX0ZBI7u9lPVm5rZLjMbM7OxkZGRglomzZTn9ZMP6ZcyLYkh3SuyB/CbRmqH5BUAXoyO1wCsanreyuiYOObDhUaWKzL9kibll3cPVOlG94rsAewFcHv09e0Avt90/KNRNdB1AP6g/H//tKvl9r3SKNQ69KwDwElclnaqrNQPufQASH4L9QHfy0ieAvA5ABMAHiR5B4DnAXwoevrDAN4F4ASA0wA+nkcbpLNOOX6f87xxbd/+3cdxz96j+MP8Quo7SNd3n92+fpEDwC6XiNDyFH7IqwrotoSHbox5rgG4M4/XlXQ6/dH5vAZQXNsXFg1z8wsA2g9Yt15sr79qBHsO11INducZMNoF4sa5Nl7npZfPFjYA7DLlp3SjH7QYXEC6+aPzdcXHbi4McXeQcRfb3Yd+uawiod3dZ97VUUmB+J69R/Hy2VeWvE6SPC6ULks7VVbqBy0FERDfc/ztdNvG1gtjUgllNz/b7ndkqY5KurDPzS8se50keXxmLicXamKjHxQAAlLmP7q4tsdpvTCmuVNOuqjmna4YYOuQbjp5fWYuSztVVuoHpYAC4nOOP05r3v0DbxvFwWOz+NXcPC6+cAAvnVl6txx3YUxKNRBLewLtLqp5pysWUy7BfsnFg7j4wgsK+cxcpvx8TTeGRAEgMGX5o4vLu+85XMO9t1wDAEsGTYHkypikyqbmYNLpopp3ddRoQkAZHhpcMgbQeJ3PvffqUnxmUj4KAOKlTnn3bitj8uj15N1zSgoo97zv6lxfR6QT7QgmXmq3ixUQP5Bbph2uXM9DcNkOX849BJ12BFMPQLzUKe+eNifv20XHh1Sci8X/tOCgX1QFJF5qV7GUtpophGUHelkmw8Xif1pw0C/qAYiXusm7d3tHX/VlB3q9q3YxG1czgP2iACDeapcmSZNCqfpFp9cA52I2rmYA+0UpIEnNp1U5u2lL2WZAp31/ew1wcam0wRXE6TNnC/tsyzwZsYoUACQVn/Lp3balTBedXt7fXgNc62zc4aFBgMDvTy8U9tlqBrBfVAYqqWyaOBDbhR8dHsKj4zd425aylDv28v62jgEA9QCX9sLq02cr+VAZqOTKp3x6mrb0u+yynwOzeU1U8+mzlf5QAJBUfBrE86ktrfo9MJtHgPP5/ZRiaAygYD4NmObBp3y6T21plefArJZolqKoB1CgKs569GlFUZ/a0uq1Q4PnditrPd6Oi3NqHqsYvngQF12woqdtNqV8NAhcIA2qhWlypobt330cC4vL/7YGB4idf/Nmby6qeQ0gi586DQIrBVQgDaqVW6/pu51Tx2Mv/kB9H2Oflj3Q0gxhUwAoUNkmIMl5WeY7dArwPt0A6CYlbAoABdKgWnlluTPuFOD7fQPQriejm5SwKQAUSLMe/dGv5RWA9vsX9/sGoFNPRjcpYVMVUMF8WPc9dL1UY2WpiW+u5KnNzWOAxKIZRh1U1XSaj+BzJZUUTwGgD3zbjCQ0vUzKyroPsC+Bv5uejC9tlf5TAChYFecClI3L5RVc0+xeaUcBoGBV34ykDFwur+Ba1p6MVJsGgQumMjv3Og10Vm25jobJmRru2Xt0ycV/eGhQhQhyjnoABUtz96mxgmLEDco2emHTz/8Oew7XKpeim5ypYft3HsfCK0snpL105qyjFomP1AMoWLdldj5ttFJFN28cPfdZLEbLn9Tm5rH70C8rORN259TxZRd/wL+ZyOKWAkDBup0LoCn5xYt7j5NWwip7iq5d+8t+bpIfZykgklsAfBHAAICvmtmEq7YUrZvBRI0VFC/Ne5mlSsaHVF5S6rHxmAjgqAdAcgDAfQBuArABwG0kN7hoiy80Jb94Se8lW77PUiXjSypv++b1GFzRemb11UhVASQNrnoA1wI4YWbPAgDJBwBsBfC0o/Y4p3K9/LXeiV9/1ciSAV+g/h5/4G2jOHhstuc79ubXWRHN+m3mouy38Vr37D16bl+CSy4exOfee3WpB7clX64CwCiAk03fnwLw9uYnkNwGYBsArF69un8tc8SniUdJKQwfUhvdipuAt+dwLfXFvtM5t75O68W/wUUqrwrzGKRY3paBmtkuALuA+oYwjpvTFz78wSbNXC5buWTSoPrBY7Ndb8bTzSzuuNeJo1Se+MhVFVANwKqm71dGx8SRxmSou799JPbC+a2fnCxVlVIeg+rdVGZ18/uUyhNfuQoAjwFYR3ItyQsB3Apgr6O2BK954DKJT6mNbuQxqN5NEEn6fQOklgAX7zlJAZnZWZJ3AZhCvQz0fjM76qIt0l0aYyBmcBPwN7WRx6B6N7O4k16nLBf9Mo3rFCXk98DZRDAze9jM3mBmrzezf3LVDul8Fz80OIDb3r6qVBuH5LEZTzezuMu86Y8vJasuhf4e0BK69j4ZGxuz6elp180olTR3NZsmDiSmf0ZLWgWUlyqfc9LnPjo81PVAedlV/T0gedjMxpIe97YKSHqXdg+CbtMYPlQpxSnyIu3rOedBs8/1HmgtoApKu66Q0hhh0uxzvQfqAVRQrztgleGC30ob7vROs8/1HigAVFCZtgHMmr4JvQufhU+zz10J/T1QAKigstzV5LFfcpmCnY/S9PyqOiBe1t5vHjQGUEFlyennsQdCtxvuSDYaa6km9QAqqgx3NXmkb5K68EC9xK9qd6uuaKylmhQAxJm80jetwa7dgnZZln0OmcZaqkkpIHGmqPRN0t3q7kO/VAqjjcaCgGvH92HTxIEl703o5ZJVpQAgzuQ5VtF88Uqa1dw6593n1Uz7rVOOX2Mt1aQUkDiVx1hFa8onDaUw6jrl+EMvl6wqBQApvW5WMyWW9wAApTAausnxl6GwQNJRCkhKr91dfCO19OHrViuFkWBypoYVXL6BPKAAWXXqAUhuXE0USqomal3Rcex1lyqF0aKRPovb60EBsvoUACQXvczqzStgrPmz+ABw/VUjS75XCmO5pPTZAOnl5EHJl1JAkou0s3rzmlk6OVPDf//v72IfO3hsNtXvClFS+uwVM138A6AAILlIO1Eoj2UgGr8naUsjVfh0pvr+sCkFJLHSpmfSzurNa2Zpu+dX4SJW9LhKWRYOlGKoByDL9JKeSTtRKK87z6TnM2pTmfVjAbayLBwoxVAPIFDt7ix7Wfgr7UShvO48434PAXz4utWlv4j1awE2DY6HSwEgQJ0qdnpNz6S5kOQ1s7TKM1S1AJsUTQEgQJ3uLPu1yUojYDR6I3//7SPYOXU89QW8qnew2uxGiqYxgAB1urPs58Jf2mgkmRZgk6KpBxCA1nz/a4cGMTe/sOx5jTvLfqZVtNFIsiqnt8QPCgAVF5fvHxwgBlcQC6+cr6BvvbPsV1pFee72qpreEj8oBVRxcXfYC4uGV7/qAi9K/zQRScQd9QAqLulOeu70Amb+8a/73JrlNBFJxB31ACrO9ztsTUQScUc9gIorwx228twibigAVJwqSUQkSaYAQPKDAO4B8BcArjWz6abHdgC4A8AigL8zs6no+BYAXwQwAOCrZjaRpQ3Sme6wy83VRjtSfVl7AE8BuAXAvzUfJLkBwK0ArgZwJYBHSL4hevg+AO8EcArAYyT3mtnTGdshUkm9bLQj0q1Mg8Bm9oyZxS3gvhXAA2b2spk9B+AEgGujfyfM7FkzOwPggei5IhIjr30TROIUVQU0CuBk0/enomNJx5chuY3kNMnp2Vnt7CRh0kQ5KVLHAEDyEZJPxfwr9M7dzHaZ2ZiZjY2MjHT+AZEK8r2MV8qt4xiAmb2jh99bA7Cq6fuV0TG0OS4iLcpQxivlVVQKaC+AW0leRHItgHUAfgrgMQDrSK4leSHqA8V7C2qDSOlpopwUKWsZ6PsB/DOAEQD7SB4xs81mdpTkgwCeBnAWwJ1mthj9zF0AplAvA73fzI5mOgORHPlYcqkyXikKzazzsxwbGxuz6enpzk8UyaC15BKop1t0xy1lRfKwmY0lPa61gEQiKrmU0CgAiERUcimhUQAQiajkUkKjACAS0R68EhqtBioSSbNyqo/VQiJpKQCINOmm5FILtElVKAUkkpKqhaQq1APwXK+phlBTFGnOu9f3SNVCUhUKAB7rNdUQaooizXlneY+uHB5CLeZir2ohKRulgDzWa6oh1BRFmvPO8h6pWkiqQj0Aj/Waagg1RZHmvLO8R9pnWapCAcBjSamGFSTWju9LvPCEmqJIc95Z3yMt0CZVoBSQx+JSDQCwaAZDPW9997eP4C2f/yEmZ2ptf67sKYrJmRo2TRzA2vF92DRxYMn5NqQ57yq+RyJpqQfgsdZUwwoSizGrt87NLywZwKxaiqLbAds0512190ikF1oOukTWju9Du09rdHgIj47f0Lf29MumiQOx6Zqqnq9IXrQcdIV0yk9XdZA31EFtkaIpAJRI0phAQ1UHebVKp0gxKh0Auhk4LJPG/rCXXDy47LEqD2BqwFakGJUdBK7qbNjGIG9ISz1owFakGJUdBNbAoYiELthBYA0cioi0V9kAoIFDEZH2KhsANHAoItJeZQeBNXAoItJeZQMAoAW7RETaqWwKSERE2lMAEBEJlAKAiEigFABERAKlACAiEigFABGRQGUKACR3kjxG8gmS3yM53PTYDpInSB4nubnp+Jbo2AmS41leX6RZ1VZ/FSla1h7AfgBvNLM3Afg5gB0AQHIDgFsBXA1gC4B/ITlAcgDAfQBuArABwG3Rc0Uyaaz+WpubP7df8o6HnswcBBRUpMoyBQAz+6GZnY2+PQRgZfT1VgAPmNnLZvYcgBMAro3+nTCzZ83sDIAHoueKZLJz6vi5pb8b5hcWsXPqeM+/s6igIuKLPMcAPgHgB9HXowBONj12KjqWdHwZkttITpOcnp2dzbGZUkVFrP5aRFAR8UnHAEDyEZJPxfzb2vSczwI4C2B3Xg0zs11mNmZmYyMjI3n9WqmoIlZ/1ZLiUnUd1wIys3e0e5zkxwC8B8CNdn53mRqAVU1PWxkdQ5vjIj3bvnn9kh3ggOyrv145PBS7qZCWFJeqyFoFtAXApwG8z8xONz20F8CtJC8iuRbAOgA/BfAYgHUk15K8EPWB4r1Z2iACnN8veXR4CER957d7b7km02KAWlJcqi7raqBfBnARgP0kAeCQmf2tmR0l+SCAp1FPDd1pZosAQPIuAFMABgDcb2ZHM7ZBBED+q79qSXGpusruCSwiErpg9wQWEZH2FABERAKlACAiEigFABGRQCkAiIgESgFARCRQCgAiIoFSABARCZQCgIhIoBQAREQCpQAgIhIoBQARkUBlXQ20NCZnalrVUUSkSRABoLG3a2OzkMbergAUBEQkWEGkgLS3q4jIckEEAO3tKiKyXBApIO3tKq00JiQSSA9Ae7tKs8aYUG1uHobzY0KTMzXXTRPpqyACQBEbhkt5aUxIpC6IFBCQ/4bhUl4aExKpC6IHINIsaexHY0ISGgUACY7GhETqgkkBiTQ0UoGqApLQKQBIkDQmJKIUkIhIsBQAREQCpQAgIhIoBQARkUApAIiIBIpm5roNHZGcBfC863Z0cBmA37puRJ+Ecq46z+oJ5Vwb5/k6MxtJelIpAkAZkJw2szHX7eiHUM5V51k9oZxrt+epFJCISKAUAEREAqUAkJ9drhvQR6Gcq86zekI5167OU2MAIiKBUg9ARCRQCgAiIoFSAMgRyZ0kj5F8guT3SA67blMRSH6Q5FGSr5CsXEkdyS0kj5M8QXLcdXuKQvJ+ki+SfMp1W4pEchXJgySfjv6//aTrNhWF5KtI/pTk49G5fr7d8xUA8rUfwBvN7E0Afg5gh+P2FOUpALcA+LHrhuSN5ACA+wDcBGADgNtIbnDbqsJ8HcAW143og7MAPmVmGwBcB+DOCn+mLwO4wczeDOAtALaQvC7pyQoAOTKzH5rZ2ejbQwBWumxPUczsGTOr6g7q1wI4YWbPmtkZAA8A2Oq4TYUwsx8D+J3rdhTNzF4ws59FX/8JwDMAKrkZhNX9X/TtYPQvsdJHAaA4nwDwA9eNkNRGAZxs+v4UKnqxCBHJNQA2AviJ25YUh+QAySMAXgSw38wSz1U7gqVE8hEAfx7z0GfN7PvRcz6Lerdzdz/blqduzlOkTEi+GsAeAHeb2R9dt6coZrYI4C3RGOT3SL7RzGLHeRQAUjKzd7R7nOTHALwHwI1W4kkWnc6zwmoAVjV9vzI6JiVGchD1i/9uM3vIdXv6wczmSB5EfZwnNgAoBZQjklsAfBrA+8zstOv2SE8eA7CO5FqSFwK4FcBex22SDEgSwNcAPGNmX3DdniKRHGlUH5IcAvBOAMeSnq8AkK8vA3gNgP0kj5D8V9cNKgLJ95M8BeAvAewjOeW6TXmJBvHvAjCF+mDhg2Z21G2rikHyWwD+B8B6kqdI3uG6TQXZBOAjAG6I/i6PkHyX60YV5AoAB0k+gfrNzH4z+4+kJ2spCBGRQKkHICISKAUAEZFAKQCIiARKAUBEJFAKACIigVIAEBEJlAKAiEig/h+No222BbIEaQAAAABJRU5ErkJggg==\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "for i in range(0,5):\n", + " plt.scatter(X[:,i],y)\n", + " plt.show()\n", + "# Your code here" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "XMXb9lTyzGHE" + }, + "source": [ + "## Model Definition\n", + "\n", + "Using TensorFlow, build a model with the following definition:\n", + "> Input of shape 5 \\\\\n", + "> Dense of shape 5 \\\\\n", + "> Dense of shape 5 \\\\\n", + "> Dense of shape 1 \\\\\n", + "\n", + "Use Mean Square Error Loss and Stochaistic Gradient Descent (SGD) Optimizer\n", + "\n", + "Use Gradient Decay with appropriate parameters" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "r32N1xK2ziOs", + "outputId": "62942a08-0b5a-4e53-a402-4c56d24fc546" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Epoch 1/20\n", + "5359/5359 [==============================] - 11s 2ms/step - loss: 10069.8164 - val_loss: 3255.5579\n", + "Epoch 2/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 823.1278 - val_loss: 285.4702\n", + "Epoch 3/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 252.7530 - val_loss: 207.0333\n", + "Epoch 4/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 158.5191 - val_loss: 112.7389\n", + "Epoch 5/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 89.4986 - val_loss: 63.2304\n", + "Epoch 6/20\n", + "5359/5359 [==============================] - 9s 2ms/step - loss: 55.4603 - val_loss: 42.9696\n", + "Epoch 7/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 41.3788 - val_loss: 35.1457\n", + "Epoch 8/20\n", + "5359/5359 [==============================] - 9s 2ms/step - loss: 35.7744 - val_loss: 33.0369\n", + "Epoch 9/20\n", + "5359/5359 [==============================] - 9s 2ms/step - loss: 33.1982 - val_loss: 30.5061\n", + "Epoch 10/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 31.5505 - val_loss: 33.5671\n", + "Epoch 11/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 30.4445 - val_loss: 28.0611\n", + "Epoch 12/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 29.5325 - val_loss: 33.2136\n", + "Epoch 13/20\n", + "5359/5359 [==============================] - 9s 2ms/step - loss: 29.1059 - val_loss: 28.0553\n", + "Epoch 14/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 28.3533 - val_loss: 38.9509\n", + "Epoch 15/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 28.2272 - val_loss: 27.2245\n", + "Epoch 16/20\n", + "5359/5359 [==============================] - 9s 2ms/step - loss: 27.9115 - val_loss: 28.3221\n", + "Epoch 17/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 27.5732 - val_loss: 29.7209\n", + "Epoch 18/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 27.6376 - val_loss: 26.4898\n", + "Epoch 19/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 27.6782 - val_loss: 28.0481\n", + "Epoch 20/20\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 27.5745 - val_loss: 28.2481\n" + ] + } + ], + "source": [ + "import tensorflow as tf\n", + "from tensorflow import keras\n", + "from tensorflow.keras import layers\n", + "\n", + "model = keras.Sequential()\n", + "model.add(keras.Input(shape=(5)))\n", + "model.add(layers.Dense(5, activation='relu'))\n", + "model.add(layers.Dense(5, activation='relu'))\n", + "model.add(layers.Dense(1))\n", + "# compile model\n", + "optimizer = keras.optimizers.SGD(learning_rate=0.1)\n", + "model.compile(loss='mse')\n", + "# fit model\n", + "history = model.fit(X_train, y_train, validation_split=0.33, epochs=20,batch_size=1)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "jmeP6vt3z0oA" + }, + "source": [ + "## Plot Loss\n", + "\n", + "Using matplotlib visualise how the loss (both validation and training) is changing, use this information to retrain the model with appropriate parameters.
We ideally want the loss to be constant over the last few iterations." + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 295 + }, + "id": "RQTNqPHm0mOi", + "outputId": "4d5b5a85-7cf7-4295-d485-6c42deb2b66b" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ], + "source": [ + "# Your code here\n", + "import matplotlib.pyplot as plt\n", + "loss_train = history.history['loss']\n", + "loss_val = history.history['val_loss']\n", + "plt.plot(loss_train, 'b', label='Training loss')\n", + "plt.plot(loss_val, 'g', label='validation loss')\n", + "plt.title('Training and Validation loss')\n", + "plt.xlabel('Epochs')\n", + "plt.ylabel('Loss')\n", + "plt.legend()\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "IVrR_vXA7kOt" + }, + "source": [ + "## Evaluation Metrics\n", + "Use the R2 Score function implemented in the first assignment to evaluate the performance of the model." + ] + }, + { + "cell_type": "code", + "execution_count": 21, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "-lOHpD8-7ggm", + "outputId": "7e20a23a-e7da-4bb2-bf72-9703281223dd" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Accuracy: 0.9980324352162894\n" + ] + } + ], + "source": [ + "import numpy as np\n", + "predictions=model.predict(X_test)\n", + "from sklearn.metrics import r2_score\n", + "accu = r2_score(y_test, predictions) \n", + "print(\"Accuracy:\", accu)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "CHqzF1OU0pBg" + }, + "source": [ + "## Your own custom model\n", + "Build a custom model of your own choice.
\n", + "Describe it in detail in Markdown/Latex in the cell below.
\n", + "Visualise the loss, as before." + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Q2MpwimsA8nk" + }, + "source": [ + "# Generating Dataset " + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "metadata": { + "id": "1XOk5hJu0oSQ" + }, + "outputs": [], + "source": [ + "# Your code here\n", + "from sklearn import datasets\n", + "from sklearn.model_selection import train_test_split\n", + "\n", + "# Generate the data\n", + "X, y = datasets.make_regression(n_samples=10000, n_features=5, noise=5, random_state=4)\n", + "\n", + "# Split the data\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1234)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "j-SQ2NUQBE0m" + }, + "source": [ + "# Visualizing the Dataset" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "metadata": { + "id": "zico16qLBNiS", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "outputId": "b5065a03-ce82-4020-844e-c18b3c9bc610" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "for i in range(0,5):\n", + " plt.scatter(X[:,i],y)\n", + " plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "sei-b2DUCJ9L" + }, + "source": [ + "# Model Definition" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "vUThXgGQCOpO" + }, + "source": [ + "Using TensorFlow,we have built a model with the following definition:\n", + "\n", + "\n", + "\n", + "Input of shape 5\n", + "\n", + "Dense of shape 6\n", + "\n", + "Dense of shape 6\n", + "\n", + "Dense of shape 6\n", + "\n", + "Dense of shape 4\n", + "\n", + "Dense of shape 1\n", + "\n", + "We have used Mean Square Error Loss and Adam Optimizer\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "aS0ROZa402Lo", + "outputId": "2d6beb6b-26a8-43ca-91ec-247c1ad45abd" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Epoch 1/10\n", + "5359/5359 [==============================] - 11s 2ms/step - loss: 11335.4121 - val_loss: 4405.4102\n", + "Epoch 2/10\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 1282.2136 - val_loss: 317.5958\n", + "Epoch 3/10\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 241.6147 - val_loss: 173.4451\n", + "Epoch 4/10\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 146.2795 - val_loss: 106.2516\n", + "Epoch 5/10\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 91.1051 - val_loss: 69.5142\n", + "Epoch 6/10\n", + "5359/5359 [==============================] - 11s 2ms/step - loss: 61.6492 - val_loss: 49.7792\n", + "Epoch 7/10\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 46.5812 - val_loss: 39.7649\n", + "Epoch 8/10\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 37.2249 - val_loss: 33.6605\n", + "Epoch 9/10\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 32.1133 - val_loss: 31.2136\n", + "Epoch 10/10\n", + "5359/5359 [==============================] - 10s 2ms/step - loss: 29.5520 - val_loss: 28.5606\n" + ] + } + ], + "source": [ + "import tensorflow as tf\n", + "from tensorflow import keras\n", + "from tensorflow.keras import layers\n", + "\n", + "model = keras.Sequential()\n", + "model.add(keras.Input(shape=(5)))\n", + "model.add(layers.Dense(6, activation='relu'))\n", + "model.add(layers.Dense(6, activation='relu'))\n", + "model.add(layers.Dense(6, activation='relu'))\n", + "model.add(layers.Dense(4, activation='relu'))\n", + "model.add(layers.Dense(1))\n", + "# compile model\n", + "optimizer= keras.optimizers.Adam(learning_rate=0.0003)\n", + "model.compile(loss='mse', optimizer= optimizer)\n", + "\n", + "# fit model\n", + "history = model.fit(X_train, y_train, validation_split=0.33, epochs=10,batch_size=1)\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "ATqspJ8bBDB9" + }, + "source": [ + "# Plotting the Loss" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "metadata": { + "id": "YIEAPFDYEuB0", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 295 + }, + "outputId": "c95c9435-9e14-4a3b-9f6e-40433ff8a23e" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAEWCAYAAACnlKo3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXwV1f3/8dfn3iwsYQdJWBSoKGvIjVGpVMSlVqtV9Of6pQp1rbXFpa2Cbb/anVZrKa21xbpgtVWLdavUpRVF67dWlEXBBYooYd8XCWT7/P6YSbhAAsnNvblZ3s/H45qZM2fOfO6A+TBzZs4xd0dERCQRkXQHICIizZeSiIiIJExJREREEqYkIiIiCVMSERGRhCmJiIhIwpREpMkws7+b2fhk100nM1tuZqekoN2XzeyKcHmcmb1Ql7oJHOdQM9thZtFEYz1A225mhye7XWlcSiLSIOEvmKpPpZmVxK2Pq09b7n66u89Idt2myMwmmdmcGsq7m1mpmQ2ra1vu/rC7n5qkuPZKeu7+ibvnuHtFMtqXlkdJRBok/AWT4+45wCfAl+LKHq6qZ2YZ6YuySXoIOM7M+u9TfhHwjru/m4aYROpNSURSwszGmFmxmd1sZmuA+82si5n9zczWm9nmcLlP3D7xt2gmmNlrZnZHWPcjMzs9wbr9zWyOmW03s3+Y2V1m9lAtcdclxh+a2b/C9l4ws+5x2y8xs4/NbKOZfae28+PuxcBLwCX7bLoUePBgcewT8wQzey1u/fNm9r6ZbTWz3wAWt+0zZvZSGN8GM3vYzDqH2/4IHAo8E15J3mRm/cLbThlhnV5m9rSZbTKzpWZ2ZVzbt5nZY2b2YHhuFplZUW3nYJ/v0Cncb314/r5rZpFw2+Fm9kr4fTaY2aNhuZnZL81snZltM7N36nMFJ8mhJCKplAt0BQ4DriL4+3Z/uH4oUAL85gD7Hwt8AHQHfg7ca2aWQN0/Af8BugG3sf8v7nh1ifF/gK8AhwBZwLcAzGwIcHfYfq/weDX+4g/NiI/FzI4ECsJ463uuqtroDvwV+C7BufgvMCq+CvDTML7BQF+Cc4K7X8LeV5M/r+EQjwDF4f7nAT8xs5Pitp8V1ukMPF2XmEO/BjoBA4ATCJLpV8JtPwReALoQnM9fh+WnAqOBI8J9LwA21vF4kizuro8+SfkAy4FTwuUxQCnQ5gD1C4DNcesvA1eEyxOApXHb2gEO5NanLsEv4HKgXdz2h4CH6vidaorxu3HrXwOeC5f/F3gkblv78BycUkvb7YBtwHHh+o+BpxI8V6+Fy5cC/46rZwS/9K+opd2xwLya/gzD9X7hucwgSDgVQIe47T8FHgiXbwP+EbdtCFBygHPrwOFANDxPQ+K2XQ28HC4/CEwH+uyz/0nAh8BIIJLuv/+t9aMrEUml9e6+q2rFzNqZ2e/D2xXbgDlAZ6v9yZ81VQvuvjNczKln3V7AprgygBW1BVzHGNfELe+Mi6lXfNvu/ikH+JdxGNNfgEvDq6ZxBL8wEzlXVfaNwePXzaynmT1iZivDdh8iuGKpi6pzuT2u7GOgd9z6vuemjR28P6w7kBm2VVO7NxEkw/+Et8guC7/bSwRXOncB68xsupl1rON3kSRREpFU2neI6G8CRwLHuntHglsREHfPPgVWA13NrF1cWd8D1G9IjKvj2w6P2e0g+8wguA3zeaAD8EwD49g3BmPv7/sTgj+X4WG7X96nzQMN672K4Fx2iCs7FFh5kJgOZgNQRnDrbr923X2Nu1/p7r0IrlB+a+Gjwe4+zd2PIrjqOQL4dgNjkXpSEpHG1IHg3v4WM+sK3JrqA7r7x8Bc4DYzyzKzzwJfSlGMM4EzzexzZpYF/ICD/z/2KrCF4HbNI+5e2sA4ngWGmtm54RXARILbelU6ADuArWbWm/1/6a4l6JfYj7uvAF4HfmpmbcwsH7ic4GomYR48PvwY8GMz62BmhwE3VrVrZufHPVSwmSDRVZrZ0WZ2rJllAp8Cu4DKhsQi9ackIo1pKtCW4F+e/waea6TjjgM+S3Br6UfAo8DuWuomHKO7LwKuJegYX03wC6/4IPs4wS2sw8KfDYrD3TcA5wNTCL7vQOBfcVW+DxQCWwkSzl/3aeKnwHfNbIuZfauGQ1xM0E+yCngCuNXd/1GX2A7iGwSJYBnwGsE5vC/cdjTwhpntIOisv87dlwEdgXsIzvPHBN/39iTEIvVgYQeVSKsRPiL6vrun/EpIpKXTlYi0eOFtj8+YWcTMTgPOBp5Md1wiLYHeIpbWIJfgtk03gttL17j7vPSGJNIy6HaWiIgkTLezREQkYa3udlb37t29X79+6Q5DRKTZeOuttza4e4+atrW6JNKvXz/mzp2b7jBERJoNM/u4tm26nSUiIglTEhERkYQpiYiISMJaXZ+IiDS+srIyiouL2bVr18ErS9q0adOGPn36kJmZWed9lEREJOWKi4vp0KED/fr1o/Z5xSSd3J2NGzdSXFxM//77ztpcO93OEpGU27VrF926dVMCacLMjG7dutX7alFJREQahRJI05fIn5GSSB2UVZQx5bUpvPjfF9MdiohIk6IkUgcZkQxuf/12Hlv0WLpDEZF62rhxIwUFBRQUFJCbm0vv3r2r10tLSw+479y5c5k4ceJBj3HcccclJdaXX36ZM888MyltNRZ1rNeBmRHLjTFvjQZ+FWluunXrxvz58wG47bbbyMnJ4Vvf2jPfVnl5ORkZNf8qLCoqoqio6KDHeP3115MTbDOkK5E6Kswr5J1171BWUZbuUESkgSZMmMBXv/pVjj32WG666Sb+85//8NnPfpZYLMZxxx3HBx98AOx9ZXDbbbdx2WWXMWbMGAYMGMC0adOq28vJyamuP2bMGM477zwGDRrEuHHjqBopfdasWQwaNIijjjqKiRMnHvSKY9OmTYwdO5b8/HxGjhzJwoULAXjllVeqr6RisRjbt29n9erVjB49moKCAoYNG8arr76a9HNWG12J1FEsN0ZpRSnvbXiP/J756Q5HpNm6/rnrmb9mflLbLMgtYOppU+u1T3FxMa+//jrRaJRt27bx6quvkpGRwT/+8Q9uueUWHn/88f32ef/995k9ezbbt2/nyCOP5JprrtnvnYp58+axaNEievXqxahRo/jXv/5FUVERV199NXPmzKF///5cfPHFB43v1ltvJRaL8eSTT/LSSy9x6aWXMn/+fO644w7uuusuRo0axY4dO2jTpg3Tp0/nC1/4At/5zneoqKhg586d9ToXDaEkUkexvBgAb69+W0lEpAU4//zziUajAGzdupXx48ezZMkSzIyysprvOJxxxhlkZ2eTnZ3NIYccwtq1a+nTp89edY455pjqsoKCApYvX05OTg4DBgyofv/i4osvZvr06QeM77XXXqtOZCeddBIbN25k27ZtjBo1ihtvvJFx48Zx7rnn0qdPH44++mguu+wyysrKGDt2LAUFBQ06N/WhJFJHA7sOpF1mO+atnseEggnpDkek2arvFUOqtG/fvnr5e9/7HieeeCJPPPEEy5cvZ8yYMTXuk52dXb0cjUYpLy9PqE5DTJo0iTPOOINZs2YxatQonn/+eUaPHs2cOXN49tlnmTBhAjfeeCOXXnppUo9bG/WJ1FE0EqUgt0Cd6yIt0NatW+nduzcADzzwQNLbP/LII1m2bBnLly8H4NFHHz3oPscffzwPP/wwEPS1dO/enY4dO/Lf//6X4cOHc/PNN3P00Ufz/vvv8/HHH9OzZ0+uvPJKrrjiCt5+++2kf4faKInUQ9UTWpVeme5QRCSJbrrpJiZPnkwsFkv6lQNA27Zt+e1vf8tpp53GUUcdRYcOHejUqdMB97ntttt46623yM/PZ9KkScyYMQOAqVOnMmzYMPLz88nMzOT000/n5ZdfZsSIEcRiMR599FGuu+66pH+H2rS6OdaLioo80Ump7n37Xq545go+/PqHDOw2MMmRibRc7733HoMHD053GGm1Y8cOcnJycHeuvfZaBg4cyA033JDusPZT05+Vmb3l7jU+66wrkXqo6lzXLS0Rqa977rmHgoIChg4dytatW7n66qvTHVJSqGO9Hob2GEpmJJN5q+dxwdAL0h2OiDQjN9xwQ5O88mgoXYnUQ3ZGNkMPGcrbaxqv00pEpClTEqmnWG6Meavn0dr6kkREaqIkUk+x3Bjrd65n1fZV6Q5FRCTtlETqqTCvEFDnuogIKInU24jcERjG26vVLyLSklUNqrhq1SrOO++8GuuMGTOGg70yMHXq1L3GsvriF7/Ili1bGhzfbbfdxh133NHgdhpKSaSecrJyGNhtoK5ERFqJXr16MXPmzIT33zeJzJo1i86dOycjtCZBSSQBVZ3rItI8TJo0ibvuuqt6vepf8Tt27ODkk0+msLCQ4cOH89RTT+237/Llyxk2bBgAJSUlXHTRRQwePJhzzjmHkpKS6nrXXHMNRUVFDB06lFtvvRWAadOmsWrVKk488UROPPFEAPr168eGDRsAuPPOOxk2bBjDhg1j6tSp1ccbPHgwV155JUOHDuXUU0/d6zg1mT9/PiNHjiQ/P59zzjmHzZs3Vx9/yJAh5Ofnc9FFFwE1DyXfEHpPJAGFeYU8uuhRNpVsomvbrukOR6RZuf56mJ/ckeApKICpBxjX8cILL+T666/n2muvBeCxxx7j+eefp02bNjzxxBN07NiRDRs2MHLkSM4666xa5xq/++67adeuHe+99x4LFy6ksLCwetuPf/xjunbtSkVFBSeffDILFy5k4sSJ3HnnncyePZvu3bvv1dZbb73F/fffzxtvvIG7c+yxx3LCCSfQpUsXlixZwp///GfuueceLrjgAh5//HG+/OUv1/r9Lr30Un79619zwgkn8L//+798//vfZ+rUqUyZMoWPPvqI7Ozs6ltoNQ0l3xC6EklALDd8c11XIyLNQiwWY926daxatYoFCxbQpUsX+vbti7tzyy23kJ+fzymnnMLKlStZu3Ztre3MmTOn+pd5fn4++fl7poV47LHHKCwsJBaLsWjRIhYvXnzAmF577TXOOecc2rdvT05ODueee271ZFL9+/evHs79qKOOqh64sSZbt25ly5YtnHDCCQCMHz+eOXPmVMc4btw4HnrooerZG6uGkp82bRpbtmypdVbHukrZlYiZ3QecCaxz92FhWVfgUaAfsBy4wN03W5D2fwV8EdgJTHD3t8N9xgPfDZv9kbvPCMuPAh4A2gKzgOu8kV7eiB/+5OQBJzfGIUVajANdMaTS+eefz8yZM1mzZg0XXnghAA8//DDr16/nrbfeIjMzk379+rFr1656t/3RRx9xxx138Oabb9KlSxcmTJiQUDtV9h1O/mC3s2rz7LPPMmfOHJ555hl+/OMf884779Q4lPygQYMSjjWVVyIPAKftUzYJ+Ke7DwT+Ga4DnA4MDD9XAXdDddK5FTgWOAa41cy6hPvcDVwZt9++x0qZ7u2607djX3WuizQjF154IY888ggzZ87k/PPPB4J/xR9yyCFkZmYye/ZsPv744wO2MXr0aP70pz8B8O6771ZPWbtt2zbat29Pp06dWLt2LX//+9+r9+nQoUON/Q7HH388Tz75JDt37uTTTz/liSee4Pjjj6/39+rUqRNdunSpvor54x//yAknnEBlZSUrVqzgxBNP5Gc/+xlbt25lx44dNQ4l3xApuxJx9zlm1m+f4rOBMeHyDOBl4Oaw/MHwSuLfZtbZzPLCui+6+yYAM3sROM3MXgY6uvu/w/IHgbHAnj+5FIvlqXNdpDkZOnQo27dvp3fv3uTl5QEwbtw4vvSlLzF8+HCKiooO+i/ya665hq985SsMHjyYwYMHc9RRRwFUD8M+aNAg+vbty6hRo6r3ueqqqzjttNPo1asXs2fPri4vLCxkwoQJHHPMMQBcccUVxGKxA966qs2MGTP46le/ys6dOxkwYAD3338/FRUVfPnLX2br1q24OxMnTqRz585873vfY/bs2UQiEYYOHcrpp59e7+PFS+lQ8GES+Vvc7awt7t45XDZgs7t3NrO/AVPc/bVw2z8JkssYoI27/ygs/x5QQpB8prj7KWH58cDN7n7mwWJqyFDw8W57+TZ+8MoP2D55O+2z2h98B5FWTEPBNx/NZij48KqjUfowzOwqM5trZnPXr1+flDZjuTEcZ+HahUlpT0SkOWrsJLI2vE1F+HNdWL4S6BtXr09YdqDyPjWU18jdp7t7kbsX9ejRo8FfAjT8iYgINH4SeRoYHy6PB56KK7/UAiOBre6+GngeONXMuoQd6qcCz4fbtpnZyPC22KVxbTWKPh370K1tN/WLiNSRRr5u+hL5M0rlI75/JujT6G5mxQRPWU0BHjOzy4GPgaqZnWYRPN67lOAR368AuPsmM/sh8GZY7wdVnezA19jziO/facROdQAzI5YX09wiInXQpk0bNm7cSLdu3Wp9kU/Sy93ZuHFjvV8+TOXTWRfXsmm/FyvC/pFra2nnPuC+GsrnAsMaEmNDxXJj/OqNX1FWUUZmNDOdoYg0aX369KG4uJhk9UlKarRp04Y+ffocvGIcDXvSAIV5hZRWlLJ4/WJG5I5IdzgiTVZmZib9+/dPdxiSAhr2pAGqhj/RsPAi0lopiTTAwG4DaZ/ZXk9oiUirpSTSABGLMCJ3hJKIiLRaSiINVJhbyPw186n0ynSHIiLS6JREGiiWF2NH6Q6Wblqa7lBERBqdkkgDaW4REWnNlEQaaOghQ8mMZKpfRERaJSWRBsqKZjHskGFKIiLSKimJJEEsN8bbq9/W2EAi0uooiSRBLC/Ghp0bWLm91oGERURaJCWRJFDnuoi0VkoiSTAidwSGqV9ERFodJZEkyMnK4YhuR2gMLRFpdZREkiSWF9OViIi0OkoiSVKYW8gnWz9h486N6Q5FRKTRKIkkSSwv6Fyfv2Z+miMREWk8SiJJorlFRKQ1UhJJkm7tutG3Y1/1i4hIq6IkkkSFeYVKIiLSqiiJJFEsN8YHGz5gR+mOdIciItIolESSKJYXw3EWrl2Y7lBERBqFkkgSafgTEWltlESSqE/HPnRv1139IiLSaiiJJJGZVQ8LLyLSGiiJJFksN8a7696ltKI03aGIiKRcWpKImd1gZovM7F0z+7OZtTGz/mb2hpktNbNHzSwrrJsdri8Nt/eLa2dyWP6BmX0hHd9lX7G8GGWVZSxevzjdoYiIpFyjJxEz6w1MBIrcfRgQBS4Cfgb80t0PBzYDl4e7XA5sDst/GdbDzIaE+w0FTgN+a2bRxvwuNSnMKwTUuS4irUO6bmdlAG3NLANoB6wGTgJmhttnAGPD5bPDdcLtJ5uZheWPuPtud/8IWAoc00jx1+rwroeTk5WjfhERaRUaPYm4+0rgDuATguSxFXgL2OLu5WG1YqB3uNwbWBHuWx7W7xZfXsM+aROxCCN6jtATWiLSKqTjdlYXgquI/kAvoD3B7ahUHvMqM5trZnPXr1+fykMBQef6grULqPTKlB9LRCSd0nE76xTgI3df7+5lwF+BUUDn8PYWQB9gZbi8EugLEG7vBGyML69hn724+3R3L3L3oh49eiT7++ynMK+QHaU7WLppacqPJSKSTulIIp8AI82sXdi3cTKwGJgNnBfWGQ88FS4/Ha4Tbn/J3T0svyh8eqs/MBD4TyN9hwOqmltE/SIi0tKlo0/kDYIO8reBd8IYpgM3Azea2VKCPo97w13uBbqF5TcCk8J2FgGPESSg54Br3b2iEb9KrYb0GEJmJFNPaIlIi2fBP+pbj6KiIp87d27Kj1P4+0K6t+vOC5e8kPJjiYikkpm95e5FNW3TG+spUjW3SGtL0iLSuiiJpEgsN8aGnRso3lac7lBERFJGSSRFqjrX9b6IiLRkSiIpkt8zH8PUuS4iLZqSSIrkZOVwZPcjdSUiIi2akkgKaW4REWnplERSKJYbY8W2FWzcuTHdoYiIpISSSApVDwuvW1oi0kIpiaSQhj8RkZZOSSSFurbtyqGdDtWViIi0WEoiKRbLjekxXxFpsZREUqwwr5APN37IjtId6Q5FRCTplERSLJYbw3EWrFmQ7lBERJJOSSTFNPyJiLRkSiIp1rtDb7q3665+ERFpkZREUszMqoeFFxFpaZREGkEsN8a7696ltKI03aGIiCSVkkgjiOXGKKssY9G6RekORUQkqZREGoE610WkpVISaQSHdz2cnKwcda6LSIujJNIIIhahILeAt9doDC0RaVmURBpJLDfGgjULqKisSHcoIiJJoyTSSGK5MT4t+5Slm5amOxQRkaRREmkkmltERFoiJZFGMqTHELKiWZpbRERaFCWRRpIZzWTYIcN0JSIiLUpakoiZdTazmWb2vpm9Z2afNbOuZvaimS0Jf3YJ65qZTTOzpWa20MwK49oZH9ZfYmbj0/Fd6qNqbhF3T3coIiJJUackYmbtzSwSLh9hZmeZWWYDjvsr4Dl3HwSMAN4DJgH/dPeBwD/DdYDTgYHh5yrg7jCOrsCtwLHAMcCtVYmnqSrMK2RjyUaKtxWnOxQRkaSo65XIHKCNmfUGXgAuAR5I5IBm1gkYDdwL4O6l7r4FOBuYEVabAYwNl88GHvTAv4HOZpYHfAF40d03uftm4EXgtERiaiyxXM25LiItS12TiLn7TuBc4Lfufj4wNMFj9gfWA/eb2Twz+4OZtQd6uvvqsM4aoGe43BtYEbd/cVhWW3mTld8zH8PULyIiLUadk4iZfRYYBzwblkUTPGYGUAjc7e4x4FP23LoCwINOg6R1HJjZVWY218zmrl+/PlnN1lv7rPYM6j5ISUREWoy6JpHrgcnAE+6+yMwGALMTPGYxUOzub4TrMwmSytrwNhXhz3Xh9pVA37j9+4RltZXvx92nu3uRuxf16NEjwbCTI5YX0+0sEWkx6pRE3P0Vdz/L3X8WdrBvcPeJiRzQ3dcAK8zsyLDoZGAx8DRQ9YTVeOCpcPlp4NLwKa2RwNbwttfzwKlm1iXsUD81LGvSYrkxircVs2HnhnSHIiLSYHV9OutPZtYx7Lt4F1hsZt9uwHG/ATxsZguBAuAnwBTg82a2BDglXAeYBSwDlgL3AF8DcPdNwA+BN8PPD8KyJq2qc10j+opIS5BRx3pD3H2bmY0D/k7Qh/EWcHsiB3X3+UBRDZtOrqGuA9fW0s59wH2JxJAu8XOLfP4zn09zNCIiDVPXPpHM8L2QscDT7l5GEju+W5OubbtyWKfD1C8iIi1CXZPI74HlQHtgjpkdBmxLVVAtXSwvpie0RKRFqGvH+jR37+3uXwxf+vsYODHFsbVYsdwYSzYuYUfpjnSHIiLSIHXtWO9kZndWvWthZr8guCqRBBTmFeI4C9YsSHcoIiINUtfbWfcB24ELws824P5UBdXSafgTEWkp6vp01mfc/f/FrX/fzOanIqDWoFeHXvRo10P9IiLS7NX1SqTEzD5XtWJmo4CS1ITU8pmZOtdFpEWoaxL5KnCXmS03s+XAb4CrUxZVK1CYW8iidYvYXb473aGIiCSsrk9nLXD3EUA+kB8OnHhSSiNr4WJ5Mcoqy1i0flG6QxERSVi9ZjZ0923uXvV+yI0piKfV0PAnItISNGR6XEtaFK3QZ7p+hg5ZHdQvIiLNWkOSiIY9aYCIRSjILVASEZFm7YBJxMy2m9m2Gj7bgV6NFGOLFcuNMX/NfCoqK9IdiohIQg6YRNy9g7t3rOHTwd3r+o6J1CKWF2Nn2U6WbFqS7lBERBLSkNtZ0kDqXBeR5k5JJI2G9BhCVjRLw5+ISLOlJJJGmdFMhh8yXJ3rItJsKYmkWSw3GP4kmMBRRKR5URJJs1hejE0lm1ixbUW6QxERqTclkTQrzCsENCy8iDRPSiJplt8zn4hF9ISWiDRLSiJ14A7PPAPvvpv8tttltuPIbkeqc11EmiUlkTrYvh0uuQS+//3UtF+YV6gkIiLNkpJIHXTsCNdeC48/Du+/n/z2Y7kxircVs/7T9clvXEQkhZRE6ui66yA7G37+8+S3HcsL31zX1YiINDNKInV0yCFwxRXwxz/CiiQ/jVuQWwBo+BMRaX7SlkTMLGpm88zsb+F6fzN7w8yWmtmjZpYVlmeH60vD7f3i2pgcln9gZl9Idczf+lbw8xe/SG67Xdt2pV/nfroSEZFmJ51XItcB78Wt/wz4pbsfDmwGLg/LLwc2h+W/DOthZkOAi4ChwGnAb80smsqADzsMxo2De+6BDRuS23YsN6Z3RUSk2UlLEjGzPsAZwB/CdSOYs31mWGUGMDZcPjtcJ9x+clj/bOARd9/t7h8BS4FjUh37zTdDSQlMm5bcdmO5MZZsWsL23duT27CISAql60pkKnATUBmudwO2uHt5uF4M9A6XewMrAMLtW8P61eU17LMXM7vKzOaa2dz16xv2BNTgwTB2LPz618Gjv8lS1bm+YO2C5DUqIpJijZ5EzOxMYJ27v9VYx3T36e5e5O5FPXr0aHB7kyfDli3wu98lIbhQ1fAn6lwXkeYkHVcio4CzzGw58AjBbaxfAZ3NrGq2xD7AynB5JdAXINzeCdgYX17DPil19NFw8slw552wa1dy2szLyeOQ9ofw9hr1i4hI89HoScTdJ7t7H3fvR9Ax/pK7jwNmA+eF1cYDT4XLT4frhNtf8mDc9KeBi8Knt/oDA4H/NNLXYPJkWLMGZsw4eN26MLNgWHhdiYhIM9KU3hO5GbjRzJYS9HncG5bfC3QLy28EJgG4+yLgMWAx8BxwrbtXNFawJ50ExxwTvHxYXn7w+nURy42xaP0idpfvTk6DIiIpltYk4u4vu/uZ4fIydz/G3Q939/PdfXdYvitcPzzcvixu/x+7+2fc/Uh3/3tjxm4WXI0sWwZ/+Uty2izMK6S8spxF6xclp0ERkRRrSlcizc5ZZ8GQITBlSjDSb0NVPaGl90VEpLlQEmmASCR4b2ThQpg1q+HtDegygA5ZHdQvIiLNhpJIA118cfAm+09+0vCrkYhFKMgt0PAnItJsKIk0UGZmMKbW66/Dq682vL3CvEIWrF1ARWWjPSMgIpIwJZEkuOwy6NEDfvrThrcVy42xs2wnH278sOGNiYikmJJIErRrBzfcAM89B/MaeKfLM78AABGuSURBVCdKc4uISHOiJJIkX/taMAPilCkNa2dw98FkR7PVuS4izYKSSJJ06hQkkpkzYcmSxNvJjGYyvOdwDX8iIs2CkkgSXX89ZGU1fArdquFPPBkvn4iIpJCSSBL17Bl0ss+YAcXFibcTy42xeddmPtn6SfKCExFJASWRJPv2t6GyMhjhN1HVw8Krc11EmjglkSTr1y94AXH6dNi4MbE2hvccTsQiGv5ERJo8JZEUmDQJPv00mP0wEe0y2zGo+yBdiYhIk6ckkgJDh8LZZwfzsO/YkVgbmltERJoDJZEUmTwZNm8ObmslojCvkJXbV7Lu03XJDUxEJImURFLk2GPhxBPhF7+A3QnMMRXLDd9c19WIiDRhSiIpNHkyrFoFDz5Y/30LcgsAPaElIk2bkkgKnXIKHHVU8PJhRT0H5e3Stgv9OvdTEhGRJk1JJIWqptBdujQYDqW+CvMKdTtLRJo0JZEUO+ccGDQoGCa+vqOYxHJjLNm0hG27t6UmOBGRBlISSbGqKXQXLAiGiq+Pqs71BWsWpCAyEZGGUxJpBP/zP9C3b/0nrdLcIiLS1CmJNIKsrGAK3Vdfhddeq/t+eTl59GzfU8OfiEiTpSTSSK64Arp3r9/ViJkRy4vpSkREmiwlkUbSrh1cdx3MmhX0j9RVLDfG4vWL2V2ewBuLIiIppiTSiK69FnJy6jeFbiw3RnllOe+uezd1gYmIJKjRk4iZ9TWz2Wa22MwWmdl1YXlXM3vRzJaEP7uE5WZm08xsqZktNLPCuLbGh/WXmNn4xv4u9dWlSzCF7mOPBe+O1EXV3CLqFxGRpigdVyLlwDfdfQgwErjWzIYAk4B/uvtA4J/hOsDpwMDwcxVwNwRJB7gVOBY4Bri1KvE0ZTfcAJmZcPvtdavfv0t/OmZ3VL+IiDRJjZ5E3H21u78dLm8H3gN6A2cDM8JqM4Cx4fLZwIMe+DfQ2czygC8AL7r7JnffDLwInNaIXyUhubnwla/AAw8E42odTMQiFOQWKImISJOU1j4RM+sHxIA3gJ7uvjrctAboGS73BlbE7VYcltVWXtNxrjKzuWY2d/369UmLP1Hf/jaUl8Mvf1m3+rHcGAvWLKCisp4DcImIpFjakoiZ5QCPA9e7+17jeri7A/UcJKR27j7d3YvcvahHjx7JajZhAwbARRfB3XfDpk0Hr1+YV0hJeQkfbPwg9cGJiNRDWpKImWUSJJCH3f2vYfHa8DYV4c+q2ZhWAn3jdu8TltVW3ixUTaH7m98cvK7mFhGRpiodT2cZcC/wnrvfGbfpaaDqCavxwFNx5ZeGT2mNBLaGt72eB041sy5hh/qpYVmzMHw4nHlmMIXup58euO6g7oPIjmarX0REmpx0XImMAi4BTjKz+eHni8AU4PNmtgQ4JVwHmAUsA5YC9wBfA3D3TcAPgTfDzw/Csmbjlltg40a4554D18uMZjK853AlERFpcszrOz55M1dUVORz585NdxjVxoyB//43+GRl1V7v6meu5rHFj7Hppk0EF3MiIo3DzN5y96KatumN9TSbPBmKi+Ghhw5cL5YXY8uuLXy89ePGCUxEpA6URNLs1FMhFoOf/ezAU+iqc11EmiIlkTSrmkL3ww/hr3+tvV5+z3yiFlW/iIg0KUoiTcC558IRRxx4Ct22mW0Z1H2QxtASkSZFSaQJiEbhpptg3jx44YXa62luERFpapREmohLLoE+fQ48aVUsN8aq7atY9+m62iuJiDQiJZEmIisLvvlNeOUV+L//q7lO1bDw9759r8bREpEmQUmkCbnySujWrfarkZF9RvK5Qz/HLS/dwojfjeCZD56htb3nIyJNi5JIE9K+PUycCM88A++8s//2NhltmDNhDn85/y+UVpRy1iNnMfqB0by+4vXGD1ZEBCWRJufrXz/wFLpmxnlDzmPR1xbxuzN+x9JNSxl13yjGPjKWxesXN26wItLqKYk0MV27wtVXwyOPwLJltdfLjGZyddHVLP3GUn504o946aOXGH73cC5/6nKKtxU3XsAi0qopiTRBN94IGRl1m0K3fVZ7vjP6Oyy7bhnXHXsdD73zEAN/PZCbX7yZzSWbUx+siLRqSiJNUK9eMGEC3H8/rFlTt326t+vOnV+4kw++/gHnDzmf21+/nQHTBnD7v26npKwkpfGKSOulJNJE3XQTlJXVfQrdKv069+PBcx5k/lfnc1zf47jpHzdxxG+O4L5591FeWZ6aYEWk1VISaaI+8xm44IJgCt0tW+q/f37PfJ79n2eZPX42vTr04vKnL2fE70bw1PtP6bFgEUkaJZEmbNIk2L4d7ror8TbG9BvDvy//N49f8DjlleWMfXQsn7v/c7z2yWvJC1REWi0lkSZsxAj44hdh6lTYuTPxdsyMcwefy6KvLeL3Z/6ejzZ/xPH3H89Zfz6LResWJS9gEWl1lESauMmTYcMG+MMfGt5WRiSDq466iqUTl/KTk37CKx+/Qv7v8rnsqctYsXVFww8gIq2OpsdtBkaPhuXLYenSA0+hW18bd27kJ6/+hN+8+RsM4xvHfIPJx0+ma9uuyTuIiDR7mh63mZs8GVasgD/9KbntdmvXjV984Rd8+PUPuWjYRfzi/37BgF8NYMprU9hZ1oD7ZyLSauhKpBlwh8JC2LULFi2CSIpS/ztr3+GWl27hbx/+jV4devH9Md9nQsEEMiIZqTmgiDQLuhJp5syCJ7Xefx+efDJ1xxneczjPXPwMr0x4hUM7HcqVz1zJ8LuH88R7T+ixYBGpkZJIM3HeeXD44QeeQjdZRh82mtcve50nLnwCgHMfO5fj7juOOR/PSe2BRaTZ0e2sZuSee+Cqq2DIEOjRI5h75GCfrl2D6XcTVV5ZzgPzH+DWl29l1fZVnDHwDH568k8Z3nN48r6YiDRpB7qdpSTSjJSWwg9/GPSLbNy496esrPb9OncOEkr37nVLPN26Qbt2e7exs2wn096YxpTXprBt9zbOH3o+R3Q9gg7ZHcjJyqFDVodal3OycsiMZqb25IhIyiiJxGnOSaQ27sGb7fsmlto+GzYEP3fsqL3NNm1qTi7tO5Xw9tYX+c/G59nJBsjcCRklkFlSw89dwXK0jOyM7AMnnMycOiekDtkdyI5mY2aNd5JFWrEDJZFm/9iNmZ0G/AqIAn9w91qmc2q5zKBjx+DTv3/d99u9GzZtqlvCeeed4OemTW2prDwLOKvu8UUqqcwqY0dWKTuzdrM+czeWWYJnlOAZO6mM7qAi8inl0e14RlVSWgsZy/dPThm7ILOESFYp7doa7dpGyMo0srMjtMmKkp0VITsrStvsDNpkRWmTHaVddiZtsjJok5FNdkY2bTLakB2t+3J2NFyvYVmJTFq7Zp1EzCwK3AV8HigG3jSzp91dU/zVQXY25OUFn7qqrIStW2Hz5mAolpKSPZ9du/Ze3/OJUFKSTUlJNrt2dailDpRsd0pKYGeJs2sXlO6u/bmPSmBH+KmzSBlEyiFaFi6X7VmOhtv2KyvZe599fkYynGi0kmiGE4k4kagTiUA06kQiRiTqRCNOJGpEo+y9HIVoxMiIQjQaIRp1olELyjKMjKgRjQY/M6KRvcoyq9cjZGYEn4xo8IlEgjYiESNiVr0ejey/rWo5GokE69E9y3vaiuyz/571aCRSvc++9SJmmO0bR2RPuQXrZoZh1T8jtndZxCL12q7E3riadRIBjgGWuvsyADN7BDgbUBJJkUgEunQJPslne/2srDxQYgq27doV9AfFf8rL9y8LPpmUl2dSVtaW3aWV7Npdwe7SCnaVVlJaWklpWSW7y5zS0uBTVuaUxrVXXgblJUZ5mVFeblSUR6ioMCrKo5RXRPBKCz8RvLIBTzO0SpVgVbfWPVz2Wsoqg5+wT70a6lv87fq45f3Kg79zZgeqEycuT1mtx6hLnbh4q3Yw37NrXdquPhe1tW2A06bjdrZ9GCPZmnsS6Q3ED/pUDBy7byUzuwq4CuDQQw9tnMikwSKRoIN/307+JLUeflLX4e8OFRVBMqyoqH35YNtrqltWXkl5eSWl5RWUlVdSVlFBaVkFFRWVVLhTWelUVDiVXklFpVPp4Xr8sjuVlZVUVEJlZWW47uF6WK9yzz6VFU6l773NK4Oy/ZYdHMcrodLD9erP3utV23HHCb5j9Xb23beGNgCvdNwNJyio9L0fha9qH8L64Tav/k9VWVinqq2waE9TwXH2LMdti1v2yn2OXcvfj6pjxTUfdyyrrrgnrfh+9fd8jwPXad8hNfMJNfckUifuPh2YDkHHeprDkVbCLJjmODWqkmCr+F9YmrDm/rLhSqBv3HqfsExERBpBc08ibwIDzay/mWUBFwFPpzkmEZFWo1lfC7t7uZl9HXie4BHf+9xdsyyJiDSSZp1EANx9FjAr3XGIiLRGzf12loiIpJGSiIiIJExJREREEqYkIiIiCWt1o/ia2Xrg4wR37w5sSGI4zZnOxd50Pvam87FHSzgXh7l7j5o2tLok0hBmNre24ZBbG52Lvel87E3nY4+Wfi50O0tERBKmJCIiIglTEqmf6ekOoAnRudibzsfedD72aNHnQn0iIiKSMF2JiIhIwpREREQkYUoidWBmp5nZB2a21MwmpTuedDKzvmY228wWm9kiM7su3TGlm5lFzWyemf0t3bGkm5l1NrOZZva+mb1nZp9Nd0zpZGY3hP+fvGtmfzazNumOKdmURA7CzKLAXcDpwBDgYjMbkt6o0qoc+Ka7DwFGAte28vMBcB3wXrqDaCJ+BTzn7oOAEbTi82JmvYGJQJG7DyOYruKi9EaVfEoiB3cMsNTdl7l7KfAIcHaaY0obd1/t7m+Hy9sJfkn0Tm9U6WNmfYAzgD+kO5Z0M7NOwGjgXgB3L3X3LemNKu0ygLZmlgG0A1alOZ6kUxI5uN7Airj1YlrxL814ZtYPiAFvpDeStJoK3ARUpjuQJqA/sB64P7y99wcza5/uoNLF3VcCdwCfAKuBre7+QnqjSj4lEUmImeUAjwPXu/u2dMeTDmZ2JrDO3d9KdyxNRAZQCNzt7jHgU6DV9iGaWReCuxb9gV5AezP7cnqjSj4lkYNbCfSNW+8TlrVaZpZJkEAedve/pjueNBoFnGVmywluc55kZg+lN6S0KgaK3b3qynQmQVJprU4BPnL39e5eBvwVOC7NMSWdksjBvQkMNLP+ZpZF0DH2dJpjShszM4J73u+5+53pjied3H2yu/dx934Efy9ecvcW9y/NunL3NcAKMzsyLDoZWJzGkNLtE2CkmbUL/785mRb4oEGzn2M91dy93My+DjxP8HTFfe6+KM1hpdMo4BLgHTObH5bdEs51L/IN4OHwH1zLgK+kOZ60cfc3zGwm8DbBU43zaIFDoGjYExERSZhuZ4mISMKUREREJGFKIiIikjAlERERSZiSiIiIJExJRCQJzKzCzObHfZL2praZ9TOzd5PVnkgy6T0RkeQocfeCdAch0th0JSKSQma23Mx+bmbvmNl/zOzwsLyfmb1kZgvN7J9mdmhY3tPMnjCzBeGnapiMqJndE85N8YKZtQ3rTwzndlloZo+k6WtKK6YkIpIcbfe5nXVh3Lat7j4c+A3BqL8AvwZmuHs+8DAwLSyfBrzi7iMIxp2qGh1hIHCXuw8FtgD/LyyfBMTCdr6aqi8nUhu9sS6SBGa2w91zaihfDpzk7svCgSvXuHs3M9sA5Ll7WVi+2t27m9l6oI+7745rox/worsPDNdvBjLd/Udm9hywA3gSeNLdd6T4q4rsRVciIqnntSzXx+645Qr29GeeQTDzZiHwZjj5kUijURIRSb0L437+X7j8OnumSh0HvBou/xO4Bqrnbu9UW6NmFgH6uvts4GagE7Df1ZBIKulfLSLJ0TZuVGMI5hmvesy3i5ktJLiauDgs+wbBDIDfJpgNsGq02+uA6WZ2OcEVxzUEs+LVJAo8FCYaA6ZpOlppbOoTEUmhsE+kyN03pDsWkVTQ7SwREUmYrkRERCRhuhIREZGEKYmIiEjClERERCRhSiIiIpIwJREREUnY/wdnwnRQoRjxkgAAAABJRU5ErkJggg==\n" + }, + "metadata": { + "needs_background": "light" + } + } + ], + "source": [ + "import matplotlib.pyplot as plt\n", + "loss_train = history.history['loss']\n", + "loss_val = history.history['val_loss']\n", + "plt.plot(loss_train, 'g', label='Training loss')\n", + "plt.plot(loss_val, 'b', label='validation loss')\n", + "plt.title('Training and Validation loss')\n", + "plt.xlabel('Epochs')\n", + "plt.ylabel('Loss')\n", + "plt.legend()\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "Jar-XjV_E-ps" + }, + "source": [ + "# Evaluation Metrics" + ] + }, + { + "cell_type": "markdown", + "source": [ + "Calculating the r2_score and accuracy of our model" + ], + "metadata": { + "id": "_nexYz9VLwHi" + } + }, + { + "cell_type": "code", + "execution_count": 30, + "metadata": { + "id": "VQUqZo0ZFH2R", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "441e3947-ba10-4c01-f23c-63f04ec63d21" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Accuracy: 0.9979618336142239\n" + ] + } + ], + "source": [ + "import numpy as np\n", + "predictions=model.predict(X_test)\n", + "from sklearn.metrics import r2_score\n", + "accu = r2_score(y_test, predictions) \n", + "print(\"Accuracy:\", accu)" + ] + } + ], + "metadata": { + "colab": { + "collapsed_sections": [], + "name": "Copy of Copy of CVusingTF_Assgn2.ipynb", + "provenance": [] + }, + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + }, + "language_info": { + "name": "python" + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} \ No newline at end of file From 38bd45f46f085615762fa552ee725306f078a13d Mon Sep 17 00:00:00 2001 From: Anjalijain123 <85122785+Anjalijain123@users.noreply.github.com> Date: Fri, 24 Jun 2022 22:46:49 +0530 Subject: [PATCH 3/4] Add files via upload --- Assignment 3/A3_200132.ipynb | 824 +++++++++++++++++++++++++++++++++++ 1 file changed, 824 insertions(+) create mode 100644 Assignment 3/A3_200132.ipynb diff --git a/Assignment 3/A3_200132.ipynb b/Assignment 3/A3_200132.ipynb new file mode 100644 index 0000000..eb32681 --- /dev/null +++ b/Assignment 3/A3_200132.ipynb @@ -0,0 +1,824 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "name": "Assignment 3_200132", + "provenance": [] + }, + "kernelspec": { + "name": "python3", + "display_name": "Python 3" + }, + "language_info": { + "name": "python" + }, + "accelerator": "GPU", + "gpuClass": "standard" + }, + "cells": [ + { + "cell_type": "markdown", + "source": [ + "# Assignment 3\n", + "\n", + " In this Assignment, we will use CNN to classify digits. \n", + "The `MNIST` database is a large database of handwritten digits that is commonly used for training various image processing systems.\n", + "\n" + ], + "metadata": { + "id": "VGHh_5UYzKpV" + } + }, + { + "cell_type": "markdown", + "source": [ + "## Importing TensorFlow" + ], + "metadata": { + "id": "JnsMbCPNzPAr" + } + }, + { + "cell_type": "code", + "execution_count": 33, + "metadata": { + "id": "HRLTw3cMwvi7" + }, + "outputs": [], + "source": [ + "import tensorflow as tf" + ] + }, + { + "cell_type": "markdown", + "source": [ + "## Get the dataset" + ], + "metadata": { + "id": "6Ji7HGpgzSPi" + } + }, + { + "cell_type": "code", + "source": [ + "# Import the dataset\n", + "\n", + "(x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()" + ], + "metadata": { + "id": "oEW3KDEvzIHL" + }, + "execution_count": 34, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "# Split the dataset\n", + "\n", + "from sklearn.model_selection import train_test_split\n", + "X_train,X_test,Y_train,Y_test=train_test_split(x_test,y_test,test_size=0.2)" + ], + "metadata": { + "id": "F_sRU9dx_mYQ" + }, + "execution_count": 35, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "# Pre processing \n", + "import matplotlib.pyplot as plt\n", + "plt.imshow(X_train[7], cmap=\"gray\") # Import the image\n", + "plt.show()" + ], + "metadata": { + "id": "rbt0WbW6sDVs", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 265 + }, + "outputId": "73ec6bc2-8ad6-49d9-dc17-ea53561e80d9" + }, + "execution_count": 36, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Normalize the train dataset\n", + "X_train = tf.keras.utils.normalize(X_train, axis=1)\n", + "# Normalize the test dataset\n", + "X_test = tf.keras.utils.normalize(X_test, axis=1)" + ], + "metadata": { + "id": "4Nc41NAf5H1x" + }, + "execution_count": 37, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "print(X_train.shape, Y_train.shape)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "NUeHTScw5dIG", + "outputId": "90fd0a50-970e-43f8-913f-bca782521f01" + }, + "execution_count": 38, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "(8000, 28, 28) (8000,)\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "## Visualize the dataset\n", + "Print some images with labels." + ], + "metadata": { + "id": "EVpQheoVqoEG" + } + }, + { + "cell_type": "code", + "source": [ + "import matplotlib.pyplot as plt\n", + "plt.figure(figsize=(10,10))\n", + "for i in range(10):\n", + " plt.subplot(5,5,i+1)\n", + " plt.xticks([])\n", + " plt.yticks([])\n", + " plt.grid(False)\n", + " plt.imshow(X_train[i])\n", + " \n", + " plt.xlabel(Y_train[i])\n", + "plt.show()\n", + "# Your code" + ], + "metadata": { + "id": "yF1Nj63Bz9m7", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 251 + }, + "outputId": "71819733-cab2-4e48-fe8a-7b46f45f7902" + }, + "execution_count": 39, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "Plot statistics of the training and testing dataset \n", + "(`x axis`: digits, `y axis`: number of samples corresponding to the digits)" + ], + "metadata": { + "id": "Rx8muKSIrKhe" + } + }, + { + "cell_type": "code", + "source": [ + "import numpy as np\n", + "import numpy as np\n", + "unique, counts = np.unique(Y_train, return_counts=True)\n", + "print (np.asarray((unique, counts)).T)\n", + "\n" + ], + "metadata": { + "id": "37kehTG_6Pi4", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "ea4757d1-c5dc-415b-8514-0cfc9aeb7c9a" + }, + "execution_count": 40, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "[[ 0 813]\n", + " [ 1 890]\n", + " [ 2 823]\n", + " [ 3 811]\n", + " [ 4 788]\n", + " [ 5 692]\n", + " [ 6 769]\n", + " [ 7 812]\n", + " [ 8 789]\n", + " [ 9 813]]\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "plt.bar(unique,counts)\n", + "plt.title(\"Statistics of Training dataset\")\n", + "plt.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 281 + }, + "id": "QZoj8t7VHdsl", + "outputId": "a0b92f83-bfcd-4010-eaca-107a40f8b615" + }, + "execution_count": 41, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "source": [ + "import numpy as np\n", + "unique2, counts2 = np.unique(Y_test, return_counts=True)\n", + "print (np.asarray((unique2, counts2)).T)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "jjWex3-VHioz", + "outputId": "e29942dc-df87-4090-deb3-d4013dcd000a" + }, + "execution_count": 42, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "[[ 0 167]\n", + " [ 1 245]\n", + " [ 2 209]\n", + " [ 3 199]\n", + " [ 4 194]\n", + " [ 5 200]\n", + " [ 6 189]\n", + " [ 7 216]\n", + " [ 8 185]\n", + " [ 9 196]]\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "## Model" + ], + "metadata": { + "id": "kWlpCWdAr8d3" + } + }, + { + "cell_type": "code", + "source": [ + "# model building\n", + "import tensorflow as tf\n", + "from tensorflow import keras\n", + "from tensorflow.keras import layers\n", + "from tensorflow.keras.models import Sequential\n", + "from tensorflow.keras import datasets, layers, models\n", + "import matplotlib.pyplot as plt\n", + "model = models.Sequential()\n", + "model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(28,28,1)))\n", + "model.add(layers.MaxPooling2D((2, 2)))\n", + "model.add(layers.Conv2D(64, (3, 3), activation='relu'))\n", + "model.add(layers.MaxPooling2D((2, 2)))\n", + "model.add(layers.Conv2D(64, (3, 3), activation='relu'))\n", + "model.add(layers.Flatten())\n", + "model.add(layers.Dense(64, activation='relu'))\n", + "model.add(layers.Dense(10, activation='softmax'))\n", + "model.summary()" + ], + "metadata": { + "id": "1L07EyQ0Yion", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "e7143d25-3449-46cd-d689-10666cae2cf0" + }, + "execution_count": 43, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Model: \"sequential_1\"\n", + "_________________________________________________________________\n", + " Layer (type) Output Shape Param # \n", + "=================================================================\n", + " conv2d_3 (Conv2D) (None, 26, 26, 32) 320 \n", + " \n", + " max_pooling2d_2 (MaxPooling (None, 13, 13, 32) 0 \n", + " 2D) \n", + " \n", + " conv2d_4 (Conv2D) (None, 11, 11, 64) 18496 \n", + " \n", + " max_pooling2d_3 (MaxPooling (None, 5, 5, 64) 0 \n", + " 2D) \n", + " \n", + " conv2d_5 (Conv2D) (None, 3, 3, 64) 36928 \n", + " \n", + " flatten_1 (Flatten) (None, 576) 0 \n", + " \n", + " dense_2 (Dense) (None, 64) 36928 \n", + " \n", + " dense_3 (Dense) (None, 10) 650 \n", + " \n", + "=================================================================\n", + "Total params: 93,322\n", + "Trainable params: 93,322\n", + "Non-trainable params: 0\n", + "_________________________________________________________________\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Compile the model (add optimizers and metrics)\n", + "model.compile(optimizer='adam',\n", + " loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n", + " metrics=['accuracy'])\n", + "# Fit the model on the training data (specify validation_split, read about validation if new to you)\n", + "history = model.fit(X_train, Y_train, validation_split=0.2, epochs=15,batch_size=5)\n" + ], + "metadata": { + "id": "nKEZ8cbO9JVV", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "d49ee10d-d711-4a2d-d1d0-f0b1243b6084" + }, + "execution_count": 45, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Epoch 1/15\n" + ] + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py:1082: UserWarning: \"`sparse_categorical_crossentropy` received `from_logits=True`, but the `output` argument was produced by a sigmoid or softmax activation and thus does not represent logits. Was this intended?\"\n", + " return dispatch_target(*args, **kwargs)\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "1280/1280 [==============================] - 5s 4ms/step - loss: 0.0344 - accuracy: 0.9914 - val_loss: 0.0955 - val_accuracy: 0.9756\n", + "Epoch 2/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0112 - accuracy: 0.9961 - val_loss: 0.1720 - val_accuracy: 0.9619\n", + "Epoch 3/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0214 - accuracy: 0.9931 - val_loss: 0.1145 - val_accuracy: 0.9725\n", + "Epoch 4/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0103 - accuracy: 0.9975 - val_loss: 0.1597 - val_accuracy: 0.9694\n", + "Epoch 5/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0161 - accuracy: 0.9944 - val_loss: 0.1569 - val_accuracy: 0.9737\n", + "Epoch 6/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0205 - accuracy: 0.9930 - val_loss: 0.2123 - val_accuracy: 0.9594\n", + "Epoch 7/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0129 - accuracy: 0.9967 - val_loss: 0.1392 - val_accuracy: 0.9719\n", + "Epoch 8/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0076 - accuracy: 0.9977 - val_loss: 0.1529 - val_accuracy: 0.9719\n", + "Epoch 9/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0114 - accuracy: 0.9977 - val_loss: 0.1217 - val_accuracy: 0.9781\n", + "Epoch 10/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0161 - accuracy: 0.9958 - val_loss: 0.1997 - val_accuracy: 0.9656\n", + "Epoch 11/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0102 - accuracy: 0.9970 - val_loss: 0.1380 - val_accuracy: 0.9762\n", + "Epoch 12/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0149 - accuracy: 0.9958 - val_loss: 0.1192 - val_accuracy: 0.9787\n", + "Epoch 13/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0073 - accuracy: 0.9983 - val_loss: 0.1511 - val_accuracy: 0.9769\n", + "Epoch 14/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0172 - accuracy: 0.9958 - val_loss: 0.0980 - val_accuracy: 0.9787\n", + "Epoch 15/15\n", + "1280/1280 [==============================] - 4s 3ms/step - loss: 0.0012 - accuracy: 0.9995 - val_loss: 0.1246 - val_accuracy: 0.9787\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "\n", + "plt.plot(history.history['accuracy'], label='accuracy')\n", + "plt.plot(history.history['val_accuracy'], label = 'val_accuracy')\n", + "plt.xlabel('Epoch')\n", + "plt.ylabel('Accuracy')\n", + "plt.ylim([0.5, 1.5])\n", + "plt.legend(loc='lower right')" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 296 + }, + "id": "gv9EZD0zLISC", + "outputId": "fe840335-9e8e-4666-a198-3c24695d7859" + }, + "execution_count": 46, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "" + ] + }, + "metadata": {}, + "execution_count": 46 + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "source": [ + "test_loss, test_acc = model.evaluate(X_test,Y_test, verbose=2)\n", + "print(test_acc)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "EQ-age-8LPqB", + "outputId": "925c96ef-f22c-460f-e065-3c328a44e061" + }, + "execution_count": 47, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/dispatch.py:1082: UserWarning: \"`sparse_categorical_crossentropy` received `from_logits=True`, but the `output` argument was produced by a sigmoid or softmax activation and thus does not represent logits. Was this intended?\"\n", + " return dispatch_target(*args, **kwargs)\n" + ] + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "63/63 - 0s - loss: 0.0732 - accuracy: 0.9850 - 265ms/epoch - 4ms/step\n", + "0.9850000143051147\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "source": [ + "## Predict some images\n", + "Print the image along with its label (true value) and predicted value." + ], + "metadata": { + "id": "ml1Kna_DuJrL" + } + }, + { + "cell_type": "code", + "source": [ + "# Your code\n", + "predictions = model.predict(X_test)\n", + "for i in range(5):\n", + "\n", + " print(np.argmax(predictions[i]))" + ], + "metadata": { + "id": "qioZul7_uiYq", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "dbf8d335-6626-4cf0-c169-bd1898934be9" + }, + "execution_count": 48, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "9\n", + "7\n", + "1\n", + "2\n", + "9\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "for i in range(5):\n", + " plt.imshow(X_test[i], cmap=\"gray\") # Import the image\n", + " plt.show() # Show the image" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "id": "ViSpXb5XLjLX", + "outputId": "01e135c0-82f6-4692-a194-93e48a90478b" + }, + "execution_count": 49, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAANR0lEQVR4nO3dXaxV9ZnH8d9PKDECKpYMEg4ZkOBFHZUqkjE0Y0fTRk0M9obARcMY4+lFbdqkiWOYi3qnGaetk5g0OY2mdNKxNrZGLuoMDGlCagx4IAzyVsCKlpMDTEMU0Bhe+szFWTRHPfu/N3uv/QLP95Oc7L3Xs9dejyv+WK97/x0RAnDlu6rfDQDoDcIOJEHYgSQIO5AEYQeSmN7Lhdnm1D/QZRHhqaZ3tGW3fb/tP9g+bPvJTj4LQHe53evstqdJOijpa5KOSnpL0tqI2FeYhy070GXd2LKvkHQ4Iv4YEWcl/VLSqg4+D0AXdRL2BZL+NOn10Wrap9getj1qe7SDZQHoUNdP0EXEiKQRid14oJ862bKPSVo46fVQNQ3AAOok7G9JWmp7se0ZktZI2lhPWwDq1vZufESct/24pP+WNE3SixGxt7bOANSq7UtvbS2MY3ag67pyUw2AywdhB5Ig7EAShB1IgrADSRB2IAnCDiRB2IEkCDuQBGEHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kARhB5Ig7EAShB1IgrADSRB2IIm2x2eXJNtHJJ2WdEHS+YhYXkdTAOrXUdgr/xgRf67hcwB0EbvxQBKdhj0kbbK9w/bwVG+wPWx71PZoh8sC0AFHRPsz2wsiYsz230jaLOk7EbG18P72FwagJRHhqaZ3tGWPiLHq8YSkVyWt6OTzAHRP22G3PdP27IvPJX1d0p66GgNQr07Oxs+T9Krti5/znxHxX7V0BaB2HR2zX/LCOGYHuq4rx+wALh+EHUiCsANJEHYgCcIOJFHHF2HQxJIlS4r1m2++uVhfvXp1sT5z5syGte3btxfnPXXqVLHebP533nmnWD99+nSxjt5hyw4kQdiBJAg7kARhB5Ig7EAShB1IgrADSfCttxo88sgjxfott9xSrA8NDRXrV199dbF+9uzZhrXp08u3Upw7d65YnzFjRrG+a9euYv25555rWPvwww+L86I9fOsNSI6wA0kQdiAJwg4kQdiBJAg7kARhB5Lg++wtuvbaaxvWVq5cWZy32fXkZvc6HDhwoFgfGxtrWLvzzjuL8za7hn/mzJmO5n/ooYca1rZubTh4kCTp/fffL9ZxadiyA0kQdiAJwg4kQdiBJAg7kARhB5Ig7EASXGdv0QcffNCw9vLLLxfnbXad/Y033ijWn3/++WK9pNl18JtuuqlYHx8fL9afeOKJYv2aa65pWGt2D8DJkyeL9Wb3AODTmm7Zbb9o+4TtPZOm3WB7s+1D1eOc7rYJoFOt7Mb/TNL9n5n2pKQtEbFU0pbqNYAB1jTsEbFV0mf3p1ZJ2lA93yDp4Zr7AlCzdo/Z50XExYO5Y5LmNXqj7WFJw20uB0BNOj5BFxFR+iHJiBiRNCJduT84CVwO2r30dtz2fEmqHk/U1xKAbmg37Bslrauer5P0Wj3tAOiWprvxtl+S9FVJc20flfQDSc9I+pXtRyW9J6k8gPgVwJ7yp7ib1lqpHzx4sK2eWvHJJ58U6/v27evasqXyf/t1111XnHfRokXF+p49e4p1fFrTsEfE2gal+2ruBUAXcbsskARhB5Ig7EAShB1IgrADSfAV1xatX7++Ye22224rztvsp6KPHz/eVk+DYPfu3cX67Nmz2/7sxYsXF+tcers0bNmBJAg7kARhB5Ig7EAShB1IgrADSRB2IAmus7fo6aefblh79tlni/NedVX539QHHnigWD906FCx/vHHHxfrJc16Kw1VLUlLly4t1o8dO3bJPV10/vz5tufF57FlB5Ig7EAShB1IgrADSRB2IAnCDiRB2IEkuM5eg23bthXrd999d7G+bNmyYr3ZkM07duxoWDty5Ehx3rvuuqtYv/3224v1AwcOFOud2L9/f9c+OyO27EAShB1IgrADSRB2IAnCDiRB2IEkCDuQBNfZa/DKK68U60ePHi3W16xZU6w3+/30e+65p2HtvvvKg+12+zvjzYarRu803bLbftH2Cdt7Jk17yvaY7V3V34PdbRNAp1rZjf+ZpPunmP7jiFhW/f223rYA1K1p2CNiq6STPegFQBd1coLucdu7q938OY3eZHvY9qjt0Q6WBaBD7Yb9J5KWSFomaVzSDxu9MSJGImJ5RCxvc1kAatBW2CPieERciIi/SPqppBX1tgWgbm2F3fb8SS+/IYmxc4EB52Zjh9t+SdJXJc2VdFzSD6rXyySFpCOSvhUR400XZpcXltSsWbOK9ccee6xYL31ffvr08q0U586dK9ZPniyfmx0dLZ+KmTZtWsPa2bNni/M2G7f+9ddfL9aziogpb25oelNNRKydYvILHXcEoKe4XRZIgrADSRB2IAnCDiRB2IEk+IrrADhz5kyxPjIyUqzv3bu3Ye3GG28szrt9+/ZivdlPUQ8NDRXr9957b7Fe0uyyIS4NW3YgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIILmZeBjz76qFjftGlTjzrB5YwtO5AEYQeSIOxAEoQdSIKwA0kQdiAJwg4kwXV2dOTw4cPFeulnrks/My01/648Lg1bdiAJwg4kQdiBJAg7kARhB5Ig7EAShB1IguvsGFjXX399v1u4ojTdstteaPt3tvfZ3mv7u9X0G2xvtn2oepzT/XYBtKuV3fjzkr4fEV+S9PeSvm37S5KelLQlIpZK2lK9BjCgmoY9IsYjYmf1/LSk/ZIWSFolaUP1tg2SHu5WkwA6d0nH7LYXSfqypG2S5kXEeFU6Jmleg3mGJQ233yKAOrR8Nt72LEm/lvS9iDg1uRYRISmmmi8iRiJieUQs76hTAB1pKey2v6CJoP8iIn5TTT5ue35Vny/pRHdaBFCHVs7GW9ILkvZHxI8mlTZKWlc9XyfptfrbA1CXVo7ZV0r6pqS3be+qpq2X9IykX9l+VNJ7klZ3p0UAdWga9oj4vSQ3KN9XbzsAuoXbZYEkCDuQBGEHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIKfkkZXHTt2rGFtwYIFPewEbNmBJAg7kARhB5Ig7EAShB1IgrADSRB2IAmus6Or3n333Ya1Tq+z79y5s1i/4447Ovr8Kw1bdiAJwg4kQdiBJAg7kARhB5Ig7EAShB1Ioul1dtsLJf1c0jxJIWkkIv7d9lOSHpP0f9Vb10fEb7vVKC5Ppevst956a3He+fPnF+tz585tq6esWrmp5ryk70fETtuzJe2wvbmq/Tgi/q177QGoSyvjs49LGq+en7a9XxI/MQJcZi7pmN32IklflrStmvS47d22X7Q9p8E8w7ZHbY921CmAjrQcdtuzJP1a0vci4pSkn0haImmZJrb8P5xqvogYiYjlEbG8hn4BtKmlsNv+giaC/ouI+I0kRcTxiLgQEX+R9FNJK7rXJoBONQ27bUt6QdL+iPjRpOmTT5V+Q9Ke+tsDUJdWzsavlPRNSW/b3lVNWy9pre1lmrgcd0TSt7rSIS5rFy5caFh78803i/MODQ3V3U5qrZyN/70kT1HimjpwGeEOOiAJwg4kQdiBJAg7kARhB5Ig7EASjojeLczu3cKApCJiqkvlbNmBLAg7kARhB5Ig7EAShB1IgrADSRB2IIleD9n8Z0nvTXo9t5o2iAa1t0HtS6K3dtXZ2982KvT0pprPLdweHdTfphvU3ga1L4ne2tWr3tiNB5Ig7EAS/Q77SJ+XXzKovQ1qXxK9tasnvfX1mB1A7/R7yw6gRwg7kERfwm77ftt/sH3Y9pP96KER20dsv217V7/Hp6vG0Dthe8+kaTfY3mz7UPU45Rh7fertKdtj1brbZfvBPvW20PbvbO+zvdf2d6vpfV13hb56st56fsxue5qkg5K+JumopLckrY2IfT1tpAHbRyQtj4i+34Bh+x8knZH084j4u2rav0o6GRHPVP9QzomIfx6Q3p6SdKbfw3hXoxXNnzzMuKSHJf2T+rjuCn2tVg/WWz+27CskHY6IP0bEWUm/lLSqD30MvIjYKunkZyavkrSher5BE/+z9FyD3gZCRIxHxM7q+WlJF4cZ7+u6K/TVE/0I+wJJf5r0+qgGa7z3kLTJ9g7bw/1uZgrzImK8en5M0rx+NjOFpsN499JnhhkfmHXXzvDnneIE3ed9JSLukPSApG9Xu6sDKSaOwQbp2mlLw3j3yhTDjP9VP9ddu8Ofd6ofYR+TtHDS66Fq2kCIiLHq8YSkVzV4Q1EfvziCbvV4os/9/NUgDeM91TDjGoB118/hz/sR9rckLbW92PYMSWskbexDH59je2Z14kS2Z0r6ugZvKOqNktZVz9dJeq2PvXzKoAzj3WiYcfV53fV9+POI6PmfpAc1cUb+HUn/0o8eGvR1k6T/rf729rs3SS9pYrfunCbObTwq6YuStkg6JOl/JN0wQL39h6S3Je3WRLDm96m3r2hiF323pF3V34P9XneFvnqy3rhdFkiCE3RAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kMT/AwW2Fi/nZ/WcAAAAAElFTkSuQmCC\n" + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "source": [ + "predictions=model.predict([X_test])\n", + "for i in range(25,32):\n", + " plt.imshow(X_test[i])\n", + " plt.show()\n", + " \n", + " print('label -> ',Y_test[i])\n", + " print('prediction -> ',np.argmax(predictions[i]))\n" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "id": "drh3JHDdPjRh", + "outputId": "ed35a28d-c5f7-4b43-b8db-d9f36dd37eb7" + }, + "execution_count": 50, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "label -> 1\n", + "prediction -> 1\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "label -> 2\n", + "prediction -> 2\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "label -> 0\n", + "prediction -> 0\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "label -> 5\n", + "prediction -> 5\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "label -> 1\n", + "prediction -> 1\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "label -> 4\n", + "prediction -> 4\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "label -> 2\n", + "prediction -> 2\n" + ] + } + ] + } + ] +} \ No newline at end of file From 7f210f664ef044b9071ca55cd6367b63a12c70de Mon Sep 17 00:00:00 2001 From: Anjalijain123 <85122785+Anjalijain123@users.noreply.github.com> Date: Wed, 20 Jul 2022 01:30:41 +0530 Subject: [PATCH 4/4] Add files via upload --- Final Task/Final_Task.ipynb.ipynb | 1024 +++++++++++++++++++++++++++++ 1 file changed, 1024 insertions(+) create mode 100644 Final Task/Final_Task.ipynb.ipynb diff --git a/Final Task/Final_Task.ipynb.ipynb b/Final Task/Final_Task.ipynb.ipynb new file mode 100644 index 0000000..53ecc8a --- /dev/null +++ b/Final Task/Final_Task.ipynb.ipynb @@ -0,0 +1,1024 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "rtI19Rt-H7Uc" + }, + "source": [ + "## Final Task:\n", + "This is your final evaluation for the project. As decided, we will be predicting images of people into three classes: `without_mask`, `mask_weared_incorrect` and `with_mask`. " + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "id": "c2CiXcHQTbX8" + }, + "outputs": [], + "source": [ + "import tensorflow as tf" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "QKDPyiZTIm1c" + }, + "source": [ + "### Loading the dataset\n", + "Make a copy of the dataset given to you in your Google Drive (keep it outside, don't put it in any folder to avoid inconvenience). Ensure it is named as `Mask_Dataset` or change the path (the variable `data_dir`) accordingly." + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "hNEMe7XsIjrK", + "outputId": "5cba9628-8fe4-44b4-9fdb-e6607a2b133d" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Mounted at /content/drive\n" + ] + } + ], + "source": [ + "# Mount Google Drive\n", + "from google.colab import drive\n", + "drive.mount('/content/drive')" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": { + "id": "8CXzo4MOJOl8" + }, + "outputs": [], + "source": [ + "import pathlib\n", + "\n", + "path='/content/drive/MyDrive/Mask_Dataset/'\n", + "data_dir = pathlib.Path(path)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "YHPHkGyDKscK" + }, + "source": [ + "### Know the Dataset\n", + "Most of the code is written for you as you aren't used to these libraries. You are to go through the documentation for your benefit." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "PzbSy-vXKjD-", + "outputId": "9b0ca462-7d7c-4c6b-e770-0ebca6052f03" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "8992\n" + ] + } + ], + "source": [ + "# Print image count\n", + "image_count = len(list(data_dir.glob('*/*.png')))\n", + "print(image_count)" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "rFHWFYj5NCVm", + "outputId": "339c3792-1af8-439b-bf6d-63a9d4e61a56" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "['mask_weared_incorrect', 'with_mask', 'without_mask']\n" + ] + } + ], + "source": [ + "# Print Output Labels\n", + "import os\n", + "output_classes = os.listdir(data_dir)\n", + "print(output_classes)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 299 + }, + "id": "fESyMw90KaxN", + "outputId": "77dab5df-1f0a-4408-f841-8cbf4ae03382" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "[2994, 3004, 2994]\n" + ] + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": { + "needs_background": "light" + } + } + ], + "source": [ + "# Plot count of each ouput label\n", + "import matplotlib.pyplot as plt\n", + "\n", + "count=[]\n", + "for label in output_classes:\n", + " this_path=path+label\n", + " dir=pathlib.Path(this_path)\n", + " im_count=os.listdir(dir)\n", + " count.append(len(im_count))\n", + "\n", + "print(count)\n", + "\n", + "plt.bar(output_classes,count)\n", + "plt.title(\"Statistics\")\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 657 + }, + "id": "HDSJ2Zk5a14s", + "outputId": "b931aa20-5ad3-4981-cb2a-e8cc696ba35d" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAIAAAACACAIAAABMXPacAABgUUlEQVR4nEz92ZYkyZElCF5amEV0MTN3jwggs7Kyu6fPmR+Yr5j/n3mYruyq7sqsXIAIX81MVYSZlnkgUUfhAA+IMHdTZRGm5dK9l+j//f/6f4J5ut3HuE+7Tx/72DfzPST5tCzXzn++xLPcltyvp/5yXV4uvRFs7GPMSPzlj21Ertd1OT1zb2F+v9/HtlEGMe/b3hovTTNznzMTqrquy7KuEfH649UDEZEJgABEIMyHw4XoJNSUQT7NpiFTiQiApzs8AYCFpXcQT7Nh7pHMaiQz4AHzCA8PjwiAiBiAJ4GImZMoIzw8HQARM7FYIogASiAyIjwiCGAiEAGRmZQgAAAliJPBBHaSEJ1JGYlMIEEJAO6ZQUgiEJIzmCBMSiwMbZzEyUQAM2tj3pBisU9nxCp07S3tddq+LqSqXRtDfOz32367z2G47ykL9772dXXAw9zTE4jkdO3Cwp7p7gmwsDZlkQQ8M0HEYFYkRSAiQYAwM6gxdU0A4ZmBTEoASYkIIMGAMERERECMyOCgBAllIgigJMrgIGQAQAJJAFECQgCIgoiIgxKPM0UmAcSUQNZDYwIIRETExJlAJDIyMxPwSERkuiCJEoRMJJLBRAAgQkmISAQSDoRnRCZnJlTSKIgIUBKiRqzJPFkmKHLlXBgYFghZWxdhEps+7vvb2/7+js0QjHVZ9HRh7fu+D/NIEHNmZMa6LIF080hildZa64uIBGCRQcQiTAIQPH16RqYIM1FjCJuZeZhFBrTuiAMOIYiAhVsXFrYEExTEzCBKZCCZ0smDgiid6/wDgLDUYScyKZMpst6FDAQjA0yZRJQgAgVxAscDYM5MINORGUAiKBL1KEBM9QAAAtelSdTJEyVleiYywED9GSW3IJBQIyQ8gWSgsS6aMzocc+fMpqyslDynD9/3237fMQwWOF10OZ/BbTffxhhmRGhNk5HuIi3CE8lC0qX1LtoCaWbDbGZoiogkKCIcEYkkJhJiQiIy3CMiGWCGgJgBpAi3JiTKIgHAnSIYFSXAEZIV14IpQaA6fwAJYSKiRERSIhOUTIkEEJmeYTji4XHYQCLrOiTqeBMggIEEER0nTln/PgGqm0ZJQJInEYiJk5CZSR4JByihlMEZzJwECmePBqhya2rpHIExW6d1WZtoBG3bjLmP3T3BHQo6XZ9bP5vFNm3bR0Y0IRUGI4gs4YkkIhZRFW1BmOZjjDHnjAxKjQSJRVpSJCVIQEiOdPfITCYoQ4kVpEqcqaqtdQJ7YpjDgjzleMsTERTOyDp3EOLnAyAoEzEiEHX6IDBlZmY6VWxCUh195vFHCaDMjLosx41gZDLXQwcYJLCMekiJ42GEIzKPAEYMEuJ67gCzEpyQAsqEpCWCSZKTJJlAlBTRVZeuLByR4e4zMiFKKtpZW18yaZ9zG8MihElEhJHwSHIPj4qlAhLztJj7mPsYHsGkyZwkIIYAQUAyCYsmYk438/Tg45U8AnET7a2pamSGJdzSnUCEpKTIyKj8AgYCUELS8fpHonMms/8M+pQJykA4EilMIMkKHYl04uOwqMJGZlISiJOIqU5eCAl2AIJIIKgiTWTW86zEgiSpz8QUxPUAwoWIIhwpmZkREWmJ6eLOYGVuosqNCEkRSQgSVu2dewc3R+77dp9mEaraGwsBaUgQU3gGwMQQDpB77HPs+zQPMGlTZiUWkBBEHm8Ki1oOc/dMIhCRMEsyZTJIVVWEEhRJEcioEoIoY2JOz8x6K+uAiI7Xn8AJiHImwOA4Um8SJUAZkWiiDE6iQBU0QAoScQSyBDHAFWUYQgQCM0BEgUGUSYkkUN2bR0ZIisxEEpgYBGICE6t5QJgiHJERQKa5bXPeDY5FVLsKgwjMxCxhRE2FuS29rStEv77dbvdtRLC2pktvyukxZ4QDfGQuEVEl5mlzznTPBDNJpkRyeqIe7RHEKTOJqKkKhUCacGcmD2Qo83paKDH3kSCPjAARqyiI3cxsD0AUdWLCx2M4/vIkIkoCIr3iNDip3lkQUpmYOUEBUCCzfkGdJIGIsh4oCExUgYeJMpGUSRH1NY4KKglMyKhSqS6DEBGO0krvKS2ZgjIpE5nwsDGnWQrA4tolaXhOc/YRyhGwvpza2j09p9u++dyFZVW5LJ2Ztvu+3baMaK23tlSmc0dazOk2KUJFmCBzcxISYVAkABAzZ4RbREyCLyzK1ISFk5SYpTNDac65+cgkS6QoQQNiFuZM1JmNJUHJINDjAdStIDAdp0PVS9RD4aRGlBRIhwcoPZHOmYIgIiHKRCADVcMwMhnJ9RcAgCOCH5GtjvN49CB6JG8GCZNQSgZF6JtzI1EigjCUckYamLV5V15Ouq4i2BM2Dduep5NkOMmJBPf3bc6Zbo3Rml6WZW0yzff79v6+szALN24cMWzadLcw92kBIoYiaNyHLsyUxMfHVIbD3YbvOyN0WRoTR4BSG/euzDR9br5vOSOYUrT3zDZ330a6kbYTZCbvhOCko6A53kcGUgmVeDkpUO8eKmFQwiwq2SKSMgSUxykHiJwyQQhCdXYIAhghxJlZIT+OYgiJcFBSJRAiIJHK1FiUkjzSXd/vW+996b0zE4KIRfu6Oi9zVT2vy3lVyUxzyricyd2XU3fEbdt+9oVNZV16axLhY9+nWTJElFh2MzObw8zCPcx8mgHVJvNRHdfbX/mQiJBM2VW6NBVlTgBMxMxU5Z1o8HAkMRNpEEdWPShEVR8ymKtkqZpRQEIkLHQUPM4EiNCRqNMiLNwjkU1AxCwMBVuCs9J6JPEiAlTeqaDBR3FDlMGWTI/fWBnq6N1AgepCEgnKqIedSPUxDdSkMpwTkoWUl859abL2vnSWxPCZMXVZ5vvW+uoe+3ZX1enmkb2rLB3K+7Dbtm1jeCCJA+lmZj7Np7m7z+Fm1eO4NpbWSYSYiI6K8VHlHX0DHV0/HbGEJICISEc4sTCxEBiBCPcMsDz+QaXJrIRMzMKirMy8z90jjhaXORmIoMzwDIMKVeYSoiBQgiMs6lIkMzKqg67iOqsdEEIyBUAMBCXIUbU/FbBBRwo4olNUCAMpMhGBSHhEeCCUU0WacldpLPz4ExHILLRC933fxuzgBJm7AgHap73fbq+327AgJkOSB7F4tQIR4VXFIXGUxut6IgZxHVQAdfs9M4grGWdVU8wCiDnMwty2PTzqH2oAlmERiWAmZj3iMKHiDBOxCIkwixDDjlILRMSUhAB+QhEsxFWG0lGggpHEVUcJZXJSZBWiSGICM5gyI4iD8HhsB/4BZOTjHQIyKaNAIkpwKiWFu43BLpwTlKQMYkYC4Z4Ih4+IjIANPy0ns/RMsAwPaT1mTM/7mNPt/f22TSMm0Rag3bx3TfADmiIQgarWPIoj/AyYx+sfdfrEyHQgiZlVWTTBc8a+2xhzjMgUEY7kBMzNwxIJJogXgsZMJCRgqfDFB4zERFmQwfG8kUiiZIYIVIiJI+EZFEkIplSAmEFoKkc9UxULSFDvB4XRnMdHoAo1YMooTC6JiSgzApkpVTARp4ZZWKa5MysHlAQKoczwCCAjImwwiFgt8mk9vd+2AGtbtjFFGsRnpu37PuY+LEC9L9JaRtpw5vCIyKx0lMSgOGJOwswSHj/BwkLQmJRVOAhOjGp6RZpZzOH77tvm4cTSPJSCMjIiEk4M4gBlpiGTiIVFiOWA0xDpEQ5KZoqMwsaibgtD5ChRKnJRBEUQgggV5SGkBdLxz+4CdVOY0rP6sQLi8gCFMlFp4biSR19dMBRzqs2ZmcEOYVJkFUSF+gYCiQgzb8zamicSao4AJ5GnuQdYLdzHHNMdYG3cOol4mCf2MSMiLI6MmH/rJz183/cISzgzaRNRZoIILV2FwyNEZGmt9wVQ8+GWc8QcCRLWRtAMuGcEuDBUTlAQgZmPpC38ALojIhAF/eLnqXAiCEJEzAkIV2mRxxkdqEIWQBc2KuvS8b4gEfVd0lOZojCFgASckxOV56sqfeS5SEqiAEGrMq2A0Jr01rqqMoMy06sx9oAQE3cVvo9hienhmTMw7juzeObwsEwwqyjA5jE9DMlm7pFRMZWSPAn5KAgS7uEZTo2JmJDM3FSW3ogdZqqsXUQlnMMxR0yrd02YBERJVO++EIkQ2COTFZRaB4VC3DKrTPx58pWJjoOshEhIqhKHBGACg5yODjayQKZIZpK6VVQRPtIjkoJUJEEUFJRMyXEUYEdTEVl4UjJn/S4iVSLtXVWa8LrIurSlqXLi59FkmmVGrGdd19OX798CfN9tTCPV9/t+uqwJ7DMCUBEHp0dEukd40hEu8ShlquBBYeosyuFBECFlEibmFGFtQhyeqcrMTMzp6eZzmnmaJTdibpVVEklcNyCSE+4iStDKrYWu4AAVEkQRlfCD6G/nHyigIZBepeWBOSR7QUuRQABgisrmYAaA8DAAQSICzuQkSAQjhZBZLS9n1AtQEdKRXNdS12Vlkd5a79KbtCbShCkp4OkZFO6n85NI22cO2+57kGAm7R5u8z5TAqIKRXo4WEg8Meacc3Kii9SboqoqmhlzWqSLcFNpQpTkQYSgDFVpjVt9AEZvYnOoLAxkxDSbc4zd1+V8Ol1EFQibU4SHOXGsp8WDjDIskAGS6lSiXt8KJpkFYTIzMonQVAGfw6JQd1IPmFlm1hgMGQinhDJYNapYZwIjIoggShHkFsIZWdMQKEkkLCIjzDOOMJiF69TAQlR0WZbMZCKuTjBRQZRZKJUoAtnWMyA/frxuY4MIZTqxi953C6bNorEEiSHJUxVAdXnJfMSyCqnMhJRgpyAVVmFCCFfqS5VclJeuogA83RjZe3t5enKn79/f77dbZgrT+bSu6+KVU8xba55gxuV8nraNgc2nRwjLgSFXeYVqZxNZrzBnBDOackYawJlNRIjCLWJGgomZhAs/EmZlkAwrfB2BdPekYwwTdX0Ar1R7TMYQkWY1ESB3SPUKAAGqopQB4sgYc4JkspARSc2VCKTU+H0bc8Z9n9OTMiAcRE4CJYTfp40EM7slAzyNiDyIWVUEYeERHpk7wpHsPkFoYIZHJMFV0Zsuiy6NVcDkiTCbxPTLn3755cOvf/3rl+/fvt3e9/V07Z0Bn2NjFWK0Jr0LUWeN1hXJxkzEOMCv9MrRmZXnlNnmBMAMYgFnpKcHIYVBwBh7eAhDqnZlKDrVAyPabTJzMhElaqqayeA6KhLhQE25LdIia8gqIlGAHI5p8c9GRcechfYqsxumEJETMhmUnJQM2kbct33fh0egSYEjFjkzEiAWaY2JkXD3aZ4ePp2J2oEM/Ex7TpmIIAZTEtLHLgoWVkETiAQhwsNstq5Pzy9Kent7//rl8/22N9XrZR3D399vSF7XlZUog6FMSRlh0+cY+91m+tTM6WlRhT7ADCUmUjCDwMzCkogwdzcgpeZkmTh+kpPATNKbiBLBI2awECcJkXp4HlM3JDMlkTAlakIWGQXeFUhCINQ8Mo+mr9AInXMnFoEmkUXCvEr05CxcuPqpbcbw9IIfkRGo1zqApgLpwoLkjD2Cwiw9RbWxUkxhFsquqk0zkhCJoAxKMIUQdeWlyaIsVSSkI6zrelpPn7982e/z27d3BlpD+pxj54xl6edzT6aI0RpFIDLSzd1imhvca+blD2TmgGMAtK7KpL110XTb3QipwsqSwPRE1a4JEGnXpTcSHXOau4hkIgqTJk6iSAgpSwGlFIKscio8EJnw4wVMZHKCH0fPSZqqLFzhKapx8AQ8HUYokJyItxmbxaykDGIiZ/gxX6C5TZDSwjVNrJeGWzufztdT396/AdxF1r5oa+ZOCJ8DbiRyPq1deT3padWmzOSJzJSlsQi/fv/x+fevYY7gdeHM2Pd3ZP722y+fPv6yrN3Sxr5xk/e3NNvXZaHwsMzISeSBRAWLOjTSLr21pXcRaqKLqu2bjZFMS2td2zSz9KCj0W1NT+d1XVZPmjbDXXr3SKA6jgwwCCAhooyYHhXaWMHHI3+8AYnHpBhMooASC7Nq04SAubCk9IxwRzKCjmE+3Xe7DzP3pCTO+s1xzJlp3zYeU0QpIgKSWfdbpbXWJ4iIWtNl6a01sxmTcyYFGHla+9rkdGqnk6gQ4FW2qbZ9+O9ffsSM3rqIHKAuNxH585/+/Nuvv7Xe9rm9vb05gnKOQZfTWQkgEY59YHo4rOKMhxOhdV2WfjmfCWBkY04Tj0RmU2ldPax3meHuoY3Pl/V6ufa+7NO2fYg0JEdalZNFF0qmJEkcIUF+0iEEJMCBrSRABYEIs7I2QJmFWOkBlldljnBDUgSlM5KSwLiPuc2i9UAyi6QRAAkzM4uCOMHIzOPWZZrtvO2SwsxZPAQwBSEZWVX/0lTgKrI0WXtrSoAkRBi9dZW8naLpufcOgJkv5+vl+RmZy3p6ejr1pd83dt/N/d4E0HXphASpKnTPbR8Wk4gDXgSg3tu6LtfLJTNsDkaCKNwpU5lVOBGq4hYgPp2Wp6fr9XIBeFqoauu572bTLT25UIX6zFyI78GbqHBCCeYkBiXzMRYWTmVtoo2gTEKiVaXVPM4D6ZnhyKB0ZApROirWF4qBQHrGg/AkALNoa8w1jubMDA/33LAvkpeuWrPZcJ/u0wjRm55Ofe3qtgMqQl21dy7MsUlo68tpPZ0/7bu13t1dmJ+eny9P1zGGZ2gDc6hkEyFkb8qE87o2aU1t6dDmYNkH45jQM4m0ZVnX9fnlJdzut3fOiGFNmYHltCx9ue9bMChJiU+n9Xq5Lstyv+1j7FXXmm9zzhGZnMEcBTgHxvRh3lqrYUwAnulRoDiJSAbMBxMpi0pTghLqBggTJziTEAelI+NgitWoCMTCmZlBVFigIcKDOT2CM81mMEsEpwshGUhqzMq0dKGYGeYeCAq3zFCV03paury976AUleXU10WIkihUICKsyxNd3PP56ZlFtrHNMWPe4fby/NSXxcwYbKtGss81Ij+9PJvZPuw+Qm8D6UxzWuEuJMKn3q+Xy6dfPvo0FaIIAd5eF2F6enle+zJs3McuxsK6LsvptBL34e/3bR8jktQCw7FZQJxUgwikmTw8Z6BxC+Gkow0KjxpuK/MxKCMIkwgJUO2FuoUfWAYRM5Iy2YOremMCHxVbZjiYSTVFke5hAdcEZQqsgTjdY8BBFpyk4C5NyXonInaPMS2Ry6k9PT398ukTS47Y1qfL+fmynNfWoYKly7Job43RIppKP60rCW1b3m4xbDYlwezclkXS89Qwpj2dJROSmzD6KueuCouZizYL2TaP1PP5vF5W9xFjX5eWl3W73cHx9HJh5pTc5g2c476lha4dk26vm9n2/fv7+30md1C7GaBnVcyABUEU6JEU6cPd9iFNhFtGTPfMo59QdndjnwwwTLi13pUoMjQMSHjRWlUATlQHUYgqUebCOJAnIhJNEQQd8FQGEhROHhzOlCqkwqr9cj5/ejldT7l2MGGf9nbb5rS+LJen83LpRLg8Py3nhZtS47bIadHTaVkX6aqwnLv11pj29FSypYVwIfCuZMzaKVJBkW1VAAeMQ8zSYtIPv4+xK9PLqZ0u1/PT1T3++Pxlu702fRJGkgdCu0b66+11u923m++vszGR5rjPMX7ct/F62+8e1CQ5AkKtEch3283hweFBZEEOnRbCqVnQU4pycTIo3WNyphIIRuQsxdmBwqPefM9wQ8AOzIkOflIRloukh4P1cpCMMhAHX9ODoImueFpP53U5rZen6/Xj8+nlKksDCPcxX9/eb9uOpNY1wkX46emJ4GNOc23tdLmc+6qNwYQgJ8rwPYIzwiOI0LQiqjIjwlSQXUUqWB4UqACmBSGUARsQvl6fXz5+kN5u7/e1t327V2tmNiPTPKaPsd3f3+/3N9BAPzEzTRv7bb5vY1gm2GMzMqGcNobnGF64LI2ZxF4sXuGavBOShRuzMEVEDIOH8NGRZWZlfgAabgnUzNQj/1YxCVPN3Aq6BIpKVDNNFCEjKAOi9ZBBhKXx03X9+PJyvT5dr+eXp9PLpa9dQLmNcTpfbvfdzPKYIvG69G17n8PmdGJt7SScBd2DRJvUMLxiKR8YvzLXiC1EhYiLdsbMANw9Ez9ev44xfvnl1+vz87Cpuo4xbN8i8PLy8vXHj/fbrTV1dxszwpFgFlVp3QEsl3W9nOa0fb/P3VV16TKTfGwxYw4fMzyZiRNino4AF0erEGocLAKR4gQVw1IYwsQkGeSehY5oUzJLEenapvl0DwqwkjBxgbkZYQSAJWveUQOhlKqVdJEmJHCl7I3O5359Wi6Xfj4v66npwlpvQpOqQ2y6R8QxEHQgzc3N3XO6ZwI5kdFVl9bc4Ac+zBBhYgJnIjOYAOaD7QoW4apWI7DvA8Cf/vwnYn59e3u/b99+vL6+37T1X3/7bVlX7HuNaMzM3YsSLMK9x/n58unl42k9vb2/3bZ3Zro8PS3reZv2+fvrvu02YAPg0IVFFJ5eyCky0pFMx4ySah6JiIMhxEwsEALBIzJTmPXpep4WIIniTzg8KYlIKCnSD7EBkqK4AEnhNb0gBiPBSQwIZRP0JuvCa+elozUjFnMfpI0aM7XWkjgXuKdHhNsYu0jLcCIJp32ziQBchIQRUrAikUr1L8QUATcHgYQjCRHmmfXtmCPIzC+X0+nC16eLR3omt+aJfdo+5uvr6+1+n+bMBERrbV2aqgpz1Y1rPz0/PQlr/96BINDHjx/X9fx2u/dGyqnf7/SGLUCIhNf0uhj/FC5CIlL4uUUQKD0AMAuzqIiIAlSj0cjUU+/LwjNyuClJMlGyZwZqkujujkwiSZAn/JgrkkASkklpQY6+0Hnh69pPi/QGFWc2gM0S8EQAXLDYQeF3jqgmeREhbc0C++4EY6aFxD033zOior6INimkMJFOxKr9aBzhmRBmUY2IyPzw8kKsXXVMW1prrROrBb7/+DHnnGN4pmhfl/OyLJd1PZ9PfWnFRHDz1jsSqrwuoiqfPnxY+/r2/n5e5ddPz79/+favv//467fttsewYUGZzKqlBmjCKpyZ04+j41ItAMLCqqINEW6WmR6kSFdlz8wMZlJiRGa4e5jb9Ih4UGxAHrAMgJtQ1bIJpIUknXt7fuovT6fLufcFrE7iJOHmaeHuGXCPBB/P0qK4WszM3AHZh2WkMLqyCvbdEZMpVeRgDR1jvKyXXYSJYGDxyExVVtWIcPfzaUnSBKk0Wnh69Nau1yuL7GOsy5pE2nRZeu/Luiy9MnlSIh95JU6XZV0+LI2fzicVUVl7++gpv3z8cD7/LvLXf/39fX+PMFBDE9YuJNxUiMksEIGIIxsQA0QirMrawqZnmhkDanPjNC/RDwuYk9IR8DgmGRUEUGRkhAfgSUxczBpwoLM+nU4fn04vL+v1uvSFwc5sIt1nmGeEm5mZI5lJEuzuZoasSRCPiLDdVfqqAINcOYSiMTkFR5hbcTgfE3OkTxJmSqbq8qFCQSACCxcznkWI9b7fbNrSe1+Wfd+1dQJNm+6e7vfb7f3dK9sn4XS5EBEhmlBf2yJw28OSCJfTkslNPyTxPuN98/u+eYCF+yLt1GrkZxEIR4QALFqIKZLAKtqZ1Sk8ySIpUiPdPb1yhAiOzpiJOQvAYT4mQFRs8GAERQizAqBswCJ0WfTpvF7O67Ioa0ZNkSrbOMx8jDmnVUSvvqweQJEXgGSi6EpMwrJFdMnLckhQKlvC49CdAOGEdEWrh4JMfrAUwr0vZw+8399AQsTTBlOeliXBZrZt2z7Gvm1zzpopMrOq9t5E+f1+i4wmdF0X7pIZ+75n2NJ7b8v9vqsuL9fLpw9PL1++fX/dEuDOS2+9N5Z0m2GWbshgLvEPp3ui5HACYRhnkgeQoUwg5XB4JlFm5py2b/twA5OqgmUzv+8zEao9I4ipqQjS5t4Dz1f55cPlt18//vbbc+9ImpkBpt3m/mbhkpZzt23bxnxAekkZOee03VprrTVV0abEJObYHQgTpyQmV+Wuita5qTIXsK4iykThCVIGWIAY+31MM7d5ezPDnDNyggQJj7i/vm5jDvNv33+UtG+MAWDuo/e25f32/g4m7nq6nj68PHvmfd9GmCAbUbjvsRGIKHrjl+fLn3/7xVN/vE8joabaVBvNAbdxd/fpLnHogoirfdmHAZOQ0lXS5+aaeKgPHwwLdzP3jGRiYgKRsET6mMasNbigjIyhsJdr/8f/9PEf/vz88fmyKGdMy1GCLsvch8/dbbcxpptTHlMHJlVtS184uXgFTsQS+xju3rutSxNgTGNOEPSgHntmFFpCpU7JqFCYRJ7pbu5Wqtci5M85PXJMv49538e0INEwryu1bVtdP9XmbhnwcBHa9v3b92/b22tjOnV5vpxbb5EZYYBIhEi7nM+//fKRuf+4jc3SiUtbwdB9iHJ1TAcdi4lB5EizmQnhgzkXBI2M8EoB5GHm6e4AuGZyIPdkViKv2e7SFwWlb+n79aT/8Pcf//f/9e9/fVkXsbAxbDMyCDvlZr5tY2zT9hkRKrrookTuwcTrsiytjzbHGJnZe2+tmc+3+010xPVc8qneKFOq7YpAZCWeylMuKiJMLJ7kFnmIuKLK8Agb+77t477t933OJGm9C3/48BKJaUbEGakiKpLa+i8dAqMYPrf7dpu7AtfziZLyXCqWJEpNJuYm8nw9E+t6mffNtmm72ZjugcbchJt4HMPFzExHZLq5Z5IHIdMjk0ij5MjEiXS3OWNaRIJYmMVLlFRqHKCIj5yGGE3z48v6n//TL7/9+nRtGWMfc0wfAUTyyHi9j9vtXgz81pbTclr6gkibnhEEVMWSmap6Pp/7smz7Nqbt+0h4oueJCazKCYnM8EghgLgkFuki2nsj5mlZ7GsvTSXRtBhj7Pt+37ZtG9OTtHdtS1/W9RyeY0w/+b6P9Bj7Bsrz+dzWvts+bO4sG+Bjf7/tEXnfZmudhZt28SmMhCjTaVXilPSc02MmYWYIQgnMJdGLgBc92Y/OoFQj1V+xHuPKGpmHm7lbFevMEACWbp6lUm+i6eY+F4nna/v1l8uvvzwJmc095t3mFmnOGkFjxrb57W5pflrWczu1viYRlSIzMcfcfTdzEWmtreu6rmvrDUzv7+9jbLebd1mEqGcrQg+oijJmcJEwlagLQyTDj74lHCCbtm9j32772DKjtaZdpfW+nPq6RuLt/f39/T1LPh459l2EVaX4O0216/XU++32fn+/fX/dbpudTqfT6ZScZMEwpHsEMhSumOQbbDI1yprjJgMBOkp6wJPMEQXrFJpWKsmiTiYyAvkoXistPPRQsOmZWPraROf2zh4vH/S3T9ffPj09nZVj83kz36YPi3DmidiNZrSgRdhPp/P1chXmMSYB63pqLG6+b5vQXNd1PZ+JyNxZ5Onpqff+/v5qdttsaNOCdwAUqZapbkAykqnSFB8EkTgK1GmzSh1zJ+L1dNK21CA2Im/32+fPn798+QLgdDp1bdraaVmbtGkzCULEoiJd1CFmtmcwO0vIqicgzT3MsibkiC557uxB72aUAbgQVCgCVKNDAoqIkJw/n1BVXweGGPAMT/xPRMLSFJIwhRslrb0J07tlS1zPy59+ffnlw7mrq3uwzZhmc1gYYRLv0T27qry8LB+ezr21/X5/f78h4omony4iaK2d13U9nYh5H2Peb73369N1Wdf1vL79+Lzdv1qnyqkHHCFg5ppSCJf+t5ghlaIdRXNyn2O3MeCpvZ2Wrn0Znts2drPvP17v95uI9N6fn5/fXl+JSVWXvpxkmWEWZtOmhQVxWxu1SIxgmrEYESUswx3hjSCUa4NcOjGPt12YtZ5eYwTNACUzKyHTLbPmvkWFZ2bWYscXC7cgngeFonQlLACT8FF9eyREcL0sHz9cL+fGmE3dLDItYmbA0YxoBo0JpGhb54z31++v37+/v353y2/rj+vahblp+/TxQ7gjY9jYxwiPtvRTb5fr1ez+/vZHeI2VwMxNtSs1ImFwJjMfvCsKcyuJZ2nh0o/sUsImZoqIfR9v77fbNj5//uIRy3Ja19P1ev3Xf/l3ZuKkcb8tp8ZMw2wbc3o4cYCDaEbGzBlGclMhoeQMCQsKZVKOtrJn+3F35Sxhr2c6YAfq+LN9pBqZS0CTiFktMungmDy4sw82cJa4hFW7TTObnBPIpeFyknVV1QQqcUyL6ZkkTXglNN9w36bt4+ncvt9fv3/9ut1uxc7/+uWmjPOK6+W8bffz+Xy+XNral3XJyNvt5sjz5ZwPQXocQlRS1UVZUJx9Kuw93CLJ3aIEKURUfOYsJiQRyMw8fbtv9/vtvs19jEyY3bfNLpfrt2+v58vptm1/+ctfzXPpjRXcVJaVpM3w+3CLZFFVstfbaWmnRTsTE5UygclFRAVKUIYcQgcgMyLM0wJFCi+WWkY4ITIAVk8qPWzgMWUhLylLAIf8QwXubpZpDKyKy8rnRYQCiH2MbR+7Z1BHO6uexThiH/u237e3t3Z/ff3x4x2gpXVmnxYeucpqkH/9/fOHD/6n1pfzWXuzOXeb9v7OlOmTiNw9w5hShKQiDyEzBBAChUV6JGck4OWucSj58qg2quh2d59uM8LzfL5u+/z8+cf9Pvpysszry4d+Pn/51z/+8pdNBMuKy/Py9ML9JNPwvu+eWFclVRtGrL0xKRUVheBCfEzQySgmhRMKLE3zdEsnCzACdFwdEJdIA8ptzSSCR7plBiUpEQQiKUgBi1AiLd2TE9eGRfH3v7387//57/fblznuI3B3Hlh5OfXTdaZu220f29Kl6/lf/se/dZF+enbzH/f7nN6Er5dLvz7poh+fPv4//rd/XJb+7evn7cu36/WynrqynNdTBr1+pe9fPl/6n05LX1XDhicIvjQR5vQ5zBxE2oibKHMQZwqLEOCWkSLt1Hs7Xbfh2xYskyQzBEK6rnOz/8//8c990e9322Pwcu7X8enXv3t7e//nv3zH73/97c8fP336jXsb+z6cWvC+mXuIkOiixBm06qKdM8P2G5OF3Yh8Xbq7vI/dI8HKUJ/T5mytadFM4QhzIzWjUjp6pkW4U+CQUHJmSZmSDnI3Exrj5YKnc1fyzaaP2dZ1BfseQd1Sxoyxjblt4cksQtTb6Xw6m/vw3Ma7zchtJN9P3p+ezv3y/MunD6fT+dvXz/C53e42p8/ttLIyH9w2VBMzTBNpnCK9ZcQc2/Ag6andwFFgdM4IY8Yi2pb1dFq5dwvT1vqyQvqf/tMv2/T4H//+433+eHsf4fTtx/ns0OXl1z//6R/+8U/cPnz5/vvnP36832d8P53X1joxR9KHj58yRgJjmpFzpjScl+XUO7G8vr6unQ2UwRTFxytLnGSWdWUVpoRTeFAZVOg02+fwRyYI50Bx2T2CKIK4XKMMESpYGv7xH/7uz7/9snTxpTXy9/vGlF3ZWcGE4W5z7Lf7fRduH55fnp5eLueLZyynpTe9b3dmAUGEr9frhw8f/tf/5R+V6fu3L9++/P7Xf//3L18+7/u9i5ZSU4gyY84JG5xIGMKUQETT5raPwKC2QnphheUssK5L06WfLn09GYQwQZCmXejp6Umnny8/np+fSHi6gSgy11M/nZ7+9KffPn767f12/z//6//5X/7pn358/5Lx9PLxpYkIxXltYx9IA9CUBaJCBCr7nEXlfF7QCIPex0Dpy0AP/Oug7Lp7IAAhIo2S7WVZNfzNJOTnfyhSmbtKBITQFX//519erlcfw8eeNnzcc5pQY86kEHKCUzrcAHp5vr48P53PF2KsSzv1tm13Zlp6u14vf/d3fzqvbY7d0n3uyGDGsrTrqZ8Xvs1ba6UXSXeDWYoUfckjhGt8nDOMcrLyQ28PVaUTL8upr2fty3CIcmuygJPYbLjH0/Us//B3HjEtzI2Ie+vL6fzh6fJyXddOr7++vH3/+Pr6Skpd4umynE8t7D63N0Kc2+nUViZhRNg2NgvbW5MzFumSnN9vJuU8QcTJM9zmZOaI9PAkCDOIlIlZfo48kCTIKDu2Q1cV3vuSWEbs5KaM89rS9+8/vtx+fAkfzCxIs2GWljzuTj7OTZbrqn09L6KI9B3Ji9LT0+lyWZemp7VfLueX6/X2+uP/+vL7fn832ynTxn1pcr2cF4ntBlVuWppN5zCCqFRSyigJX+/k6UkREQxiFWFpkiBuxQJUUr5cEayLRbLc9smUHz88vbw8JyiTPCNqIoCc8/blj33fd8n5D3/36/j49Hp/A/K68IeXy7evf8D2zJkTGZyESHem4OCMtbMyt2yWsbStMTpRCCwphDMZIC7KSul2A1qlJ7McFPYQhCMOSndEMHP5HxWP7HKil8vpui4zlhM/A/P2frfb3bb5vsVw7NN991VoOV/6emJKG7cx3qKqXKaube18Wltj2t5fz6f+fD1/+E+/XS8nQr7++Pr185ft9rbvNwa0pMTwMMsIRiqJSg26XUWwCFlu080toNVeHjrISI9ghPZ2bVc55bAISH79TuIg2fZxu90ioL0JS9jc7/c//vJv232bYxDxcl6bsmJs2/b6dS7qn57PEvf9PpHDd0qCCJZlvZ47RbJPs5Dse+Z57UuX3cwzk2JpvOgKwMzGpPFQjWpEhDtYVCQJZdQU5vkwQqOEzWlzIIKJXp6fLuf1ej7p+knpiXNu+7jt+23ENI6UGRiGcAhLsnz9+nq/79tu0wzMjbRpdiUldKXr+dSarIswxdjvMcfc72uj0/Nl39O274tSY6bM8Ik4+CxE5D48vPV1UUbGVrPTtCSmIFUu9w33SSZgJREVWGT4TLgwWMgGMi0DNUzbbR/3t+dzf7ksbobIfurny5no1zH3jFzXvnSRONlJ1kXWrghnwWlpi7IAej3ft0GuBnq6rOdTvw0P8/KoM3MIcQk7H3I1zeIClLCsHOmOyBNUfoIcTNyVQ1kzr6e+3d6+f8Wl+cLmtqnqqUtvnfQkeiJpw3K772Mbw+L0p19u9/32ftvHHoCISJPe+7K0pbOyN+Hz0i5Li5i77YjJ6WGDw54vJ5t0Oa9L020mwrk0jQ/NV02qy90k0hOgCJsmsiyreiYi3WaSUMAzS72xLkraRPr5dDpf1ky01oC8nfS66trbsi6UMccehOv1ejotHr5ve/oA4qxXVT6fliY0xx5pTEh3TlxO3WwacF7bh6fLy/V+2yZTQnTMDHNlDhbL4AO1Ci2xoLmVao+gnnFomglUAxObSFOCEl3P69zu252eW2uMDGuiDTTcM2ejrtK6yELLzjSmQ/tlafdFh826do4UoaVzX6QJlF1ywgkxOOei8KTX2x0xPjxf5y7PT5dFOObdYki5MmQQkYoSURwDJKNM5oOFDQ5mouTkiqbBlF2EWJvHCRygSDTVdbke4onIy6L08TnmXJemTNNGZKzrSZXNzFsCigyg995aV5/jTpZBRGkj0l2FMyZlru30fDm9XE4/Xu9MIX3xlPNp1aae9LZt9H573819amaCuFrFAiPqATyAi4yIMUba7Im2yNP1clr7de0vT5dLg51kXZfp/vq+3ffdE4JY1uvluuZ5nebf3+9owZBzUma6Y7oRU+tYV1lUkZFz2/wePtMGM4vk2pnQni6rN16bIkwY1JuoSJM0EGlNsMP9QCBQahUwAWlhTEUDIaHyRVNduBHLMN/GHGPW1C8i3IyApfWlt5jZGCqMZWWhpuo2ZxgWqHZCAo0JEe4+OCYAYZYuPjPdkJ6WkKHSLqflelpAU7qeTk+tr8k0Zn758cMjpt93D933XVSIm5Ua+RB+lQEnImLOycxJku6907r089Io/fb26uJKsXQRpGRw6WGVFWsjiQzz2clTyuxCSYSIzQOANlHhcBOtUtot0sLNdkKeF+2tLyqi69zu99tbul3Py7IsS1fnmZ7EHEgRtE7di49sBNLWksgLwEgWbUC6DXNjXVpfhHFe2nlZyoPQzMw5MzmTbKwKRjCCWVSYyVrHouIO9xnhVEzJzE7ZzkvRO8KdMyJj7f2+b7f3N+PTedWn07KbEeXz9fzy8eNu9nbbb9t7F17X3km0tSYqBWUlkIcdHRGV+omY2G2UG2M5Wml5rUp2hRJyjiTuKljgQZwR837wicLWxiriXqWWZubkdE9KCLI1Aowywy1jMHtnNJWlNcpwG8yEDCFC+W5Qlptw5mGQGFHK9yIogtIzRJS53HuQhCAWFKXMPWxmlM8JIYkiGNH5YGoKQamY+2BySucgIDKc3MmN08vrAUgumzUqq4kMJiRrESxYiIhHEjmFBRAxhHNtglN7P/d1bbdhu7uWFpYRfEiPGIe1DR3qZHD4RObDKpi1sSo3dhXuTG6TWVZtTdqYYR4xNiMnsABLbxYcDGIm5kDwxEQhCiQpiUQGpzFDWFi4N1mbzrG/3fcUoTThZOJ2qCjj0Td6ZEbQI2AWLTkynbmziHlwBGdwNZ2ZERbzcAeUkhNTqiSVpY2wEBDzcJTIQAYFZRrC4U5hBxcVDxspoAY4EM48pOPMZQ0q5Z/amJPJ9vuPr5/bsmRSIzQmoWSYRkRYJoi5bB0e9nJ49GL0ky6dpahurakkc5bTpNvkpKasUOSsUS8xMQkI5JAsl7uDKaYIUBTcBPcyPJLGItKofAiT4BmT0or4KpwlaXvIrx9zi/prM4AsYnhxqpTBTO6JMEoVKnOzLC9vAoGYMrRueZYfmiiDCZ4o3UNGHKxwuKQTnBBlj5sP6TMAzmKDcabPOf3wiE5masrr0k9rl9Yi5vfPv5+fn1UaxWycjTEZ6u7pYFUWKTLvUaFWRPIogacoc7gqFX2ByDIOYrcyVXESGUKpVSqWh1Gm2TiqjJpdJVFZRR+ywGBCE2lNmgqDMt3nnPvwOZQRYRnJlEJctCUiiDCSAsgyxWMI1ylQuQZwmV25JyjdOVOYmKi4FZ6B8vNzQhmYgJIQQkEUEQ+ubTKSJAng8rE7THwoE5FUvMEyqg7PKGJAhHt4BChYeF3a9bJqX8yCMjtDGYvSqcupi8G1HEOkfEoQkVw2s8c1JGTWmTGTs5ColIG1mVlOB6tQ+UdEAEhVZhJmIcAPoMkjwj3MHSinLCUwMbUuRBAioagxioe7DR87hzPRKN02DockyhCIiKJk5YRwHFMvkeKPEh0Wz+TFrQzK1HLoEE5mcz90DxF0sC0eOaTcgvPnNK3cqolJcGiHDxldKVcsIwJhaZblWQ4AyXl48pMILb1pa5dz//DyXNRo2dpucdvnsKFNGwmDOQFmzjhcc36Sio7//m3cxABFhrl5TGdW6agvk0CdpoqQpCfgUtY3ZUGQXqaI5aDKLOell5lcuLsbwiM9I5iShPJhKkZaDKDD25KZiaWccJjL+oREBD/NlQ6r4pJPBaVXjOrCpRo/ZlMRaRYBd8PBvsHR0BGkjHOYtczGUMIKPyzmshSFdJBKq7RTcTDPrBlkjbjKY6s1WZaFWTIQoNs6lq4qrKUrDqSVFCD4EYF+uvtQRpoXE7rsnxAeZmZspqIHw5mEmYNBeiD4WUa3REQOCuWeHElHuKt8VXeYqZzR3EZGFJ5DIMugCE404d5UleuJVHGQVO5XwX9zIarDT6opAqHcy9MjPbi8y46io4g5EUox2YWAYFAyzXn4JStDlISljCgy04+sWG/Fz0D98B3F4cIoElKY/nFTgPSwYcjeV1Fde1u7NgUjdJolkUdMswABZbqGguiqIPUMs+BEFQdFUY7KHkFuU4W5NWHNpLI/LtKVu1NAlIhIiJQ5ibU1USlx1r7dtakIi6hlZriFHzetVMmZmSkirXdVMfMaBIqSsppHUpXOx4aHGmdzLYGoLx8Z5UkfzId9XNIxga2hZUUt4WLhKMpr67hY5f3mkZEUB/Mu43D9ysN9gP6mdK8eilJAUS8mKN3nvsF96cvSVAStaTGxdNr0zCI1RHIgfho+o+wOA+a1/wP7yBK9hHtnKT5+WJiQtkbSiA+HnMiIwyQK5CXwkt64HP8ByihHhrQRsjRiAqXoItr2fZ+7IdzGiIBqE+3EmignZhLtCcwx/bArKYcKUNSgr6YFUFZnYuKMDDewMJowjtM8bP+r5mY6mh7SVnUIuH4XgQ7bTQ/zdD8C0AGplaVASHElp+VBfoUAHkA4pYvIIRfzqXxllfP5vPR1WmgGZhiYQBKZEaighLIgTArALJuykNzvydLH2FvOp1Nbm8z9lZFkGEOIRLSBKVFuFiRZhk2RQIngWNQz5/Q5HZBlWe7bPVOn5313FVmW0/fX7X7b0i18XtZ1WS+tnxIyPZFMotK6TbvvM5krGgkngRxetik2J3MnbZ34UayWR1keNTvjSMMAURUAOCggiKgCL0BHkX8kjPIBOdQWACHhhgiKEGRVJiAmh5SMLhxuAl/7uqyrm899i3Cwnk/Xy/XF4t+0xvPV/DI/GmAWYilybhVJmQg3NLgHJSfYw9yNKEu7CMDD0yghcRAJCcweUTIOAleAKusE1aOpioQXYy9wn7abv9/3Mb2rtNa4d2oLtYVUKdJsuud833rvl+cPEW4WNPbI3XyYZySrKJg8g8GJ4+OJCD0Men9SdB7VwOFjeljpgoMkjh0L9JPkXDYxYK63v0q2fCz6KWIVQIjyR01mqMqyNCR606W1EN2LeOseSeWzpA/+bg3iDzE2Dm4Hjp6TUMIPS9nH8AuD4O5OLsQEQQ2k3GszzYN5d0TGRzopIUVkEMo+IckRATILUJrnto3h4/X9DkC1k7QgsVBPEWqoBA8WYU+MvZjdOUbcN993s3oBwcoIGBOL6tFyEIon4gX0ZtKxGaaKfHDZaVFyBhETpDBXOhzXHj96GC/hINQeSHiEI5PKYj8yMsuaWrQ192Si8kG93e73+x3qTp0OxXO1SBERh10wspQCAA73rXJo9IAHbve7PS3ZYJHDfFUOoOTbHmAEQXA8hONxlmcwkQBcNVsSiDWpcqEMDyDNcR/zdrvtw9bzSZYTSwORQ4ZzzjrJptr6aQ0P27YAe3pQI0kSIKVyoLlP21Wl4XgBgtkjyCwykUnF2C/fMRzBh5mFgeByUIooq6Wj7vlZ4wayFj75z0dz+GARP4xdIxwcIMhBZPJqz263G09HO6Ffy7xZRThRa3YeNdXfni4i0tw7a6X4iPj+/dv9+eUq6RzTowlzBKhsDo9Qc9h8RP50BhNWEHvWlpw8FlxwYWA63SJzBg3PfSbpsl6el/MVTsScrDOaTaq70xPUWKRRYx/TMCEqy9p4ZKmuKWy7hVsSiBx8LHQqmWCmC5iSWCBAHMsw+Lj0tQfoqMQzHj1a4NhF4maOoiocXoAHsEHlsl650yORnkTxsPHJTE/3Off0ICfhpZx7tYk8pmNlbUNJ5VZWU/pjGValiYj89vXL60f9tKypZZlNhR3m8bNCeWxsyQgA0g8rXpS+jrhmPcXR8CAwzz3sWLWkKW05nU6XZ5J+33ZmlqCWRFRWwqHm97kty0LE2xbmYFYiSlHoTHeCQybS6pcc34UOA/nwDIry/CyH7qTaDfY/u+mh7LUzy9M8Ig6dnblHPYCfvWrdcgYSFpZQ4LB850xlDhFiWVQMaCIhAilsEGapjFQpMIkd8MPYG0lIYgWiBPORJT7+9v37+9vZPnYQR8AiyptLCg+h5FKyFz8HkMzaegaCV1Y49lMhIoc7UCplJyIi5dbbsia119v++Y+vRKLS17WfTifVBtAcuY39vjtA+36PoNZYtSXIg4m1KSPNYaqlkiuX3YfRYRZAEpIoa6+DikPsnBRcWNCRqaOEB6Vrr4x4zG+jmOREcfQgFBHTnEWOFhBFKCamLsLL0ibzeV1Se2o3EQ8fNpQ4GaTHrDgR5A+kNQ4HX/IwyhRGJn58H2+3u7sTCZJ8RnTxpOkJCqQfHNQAJ4HIvCaIzJRJDBIWQh57BvcZrGKR00JUUiRFgvm+j+/f3z9/fRvTwuJ0Pn369Onl5cOyNCRZmESK0Iwqz2g3N5vu3loXKUuwxpxUYo7DNxkR+WBuEeDJlJHEnMnIOPj8Ze2YkRFegeNwsAJAgYf5FQgHpbyswzkDbrWirYi5mREFRUgxuSOEU5qGNkuaY27bpshQUSeE1Uo4yqiTSmQU1SnDkV5x5scrtq2cYDWJvDZ9JLmjIB8BF7wOlhJRR2a5D7Ae/ys3w8DYzVdtFrBIAiUJUiLp9e32+nazwOcv3398/7GeVnNoW8ECYE5jTlWx6WY+xvjx+uPbt29jH8uyPF36//aPv3z6cPK5oYRQ4WVE/sDIjrqz5k8UcSAntY3pwNvKYDMemg8Ah6Y2CwPNx27FRAb8YM8GEyVzwsLNUObzyZQj7fZ+H9t9bSshM3yY3e6hQgDy5w4IzSqEa6vC4fToGSyk4Nj9dMH9ftv3kc/LmHbqXNoAEc+gYAQRU9R+o6Rg1TjAyZQkIUrAMqfHMBtzesSc5p7FpSfmbdvf32/m+fzyUdup9T++fPnyf//3f9mn/ed/+M/X63XOYIF5fPny5Xa72bS397cvn7+8vb+v6/Lpw/X5yoteGF5rE1trRUmLjGI60eEg7RE4zrRSFz8qysfkpy50PBy6wz2Z6rIfuSGyRm3uScQRCSJWSfMMK7heGUgXysvl7BFzjh9vt69fvu8T2lQiEg8IrZr92qxWUoKMNCJlaaCEf/woY9/utz3iOqadltU8PV0BCEtKpaUomJfQWR4IHlFktfJjxj58TsuI+5iRyUThHgQW2sckoPe+rOuHD5+enp7c4/fff//69dunT5/W08nccuY0+/Lly/v7rbbGLevqEeuynk7n+7Z9/bpfzou2rqq99wfTHcWnezyAeLBwEvXtIxBW6NIxGCm8xMNrs5O74Jh+uMdhtHI8DBBrCRxVtGIYMZ1Ulq5EvPZlkv7x5XWz+fr6+v37dwDalCMInlmWi4dCifOxLTQRCnCAybWjSae8j33M6SIaSYxEIkD+2BVUsHwVAzNqWw9nkkXBWiiJAECqyqIeYemeXjf9vPTzad33fH99n/s2p//6y8uH58t6OnWl7f4+5z7GcI/LqZ9PSwWIjx+eEinSlkZrGyA7QIgHiZGIRVAqS6BUjPSzxf2JwFdBFAk85ETllgEwELWwp7qHohfGsdy14G+uIHmErshSFgnlujaR5TZsXdrNvfBaJiizJCAZ1XbRQ6NUpS57OLIdlyNPSqdlobHdb9u272dpGfBaYhfsh2dAzRRrqkPhUS3PYb4fsIjpf4NziSDMGR6RnFkrM/q6riuExMxV+Hz6SEQeMcZ2v7+Psa3r6ddfPl0up76sQBKoL8vSWyaF3b9+/he7f+vS5Oda2sxH3VmwGyKpzKuA8l0LDwsqT7DHm4+jCfg59qda/Fo5IA5Z14GCH5mCjz9di7CUhCHMS++9L4ZcuvLNy6eIAM30jCSQ0LFMp6AI1J5ALm95hIUSlgWn3mPI/bbt9/3p5Ry1maDeXcqgjOp0jpY6MoXL5T8RmWbulqVELlJbfTtECJKKy9OkCa29n/tyu21zzHpB933EDMtdOQgz7Ka8rK1akFwa1s7mvm37qTVqT2vrS+siwomsDtwzIpQFBXfnTzwh/ZCwlGM6HZB/eT9XIqg+Jx6mJaW0/in+egCr+Xh/swafYCFaVBYhplCiRRthm3MvRaaWmL982oXocEFhZpSXZXqSIGZMAXVGF53M475tt40+PNVmJDpKsqi5bZAflTD4sTA2OdKjHkDFScpEzdyrfS2/GRVmJUIIvGtD4z0R7kBKQ386+blnpqqeTqfLSddOZTQo7OQ7hyt7Py/KTYSlrL8KDY1Mf6gPgcPtoFiaR6AiIEkpUIhdrQs+Ztk/eQAVlqo/c69iEccP8LGXNzzyMSFvymVrinBhWteF5d3M5sxIqHuEA5IsWf24CLeaARMyEclwIw9JLyinqyDH2Ec89H/FCOEMdspjrdPR9jAjMuBBSA+4uxetIR7j/whkMNOi2mt7R7qZuc0ES2aDOwxEbW29nYg53FV1Wdayl2chgMItfHDivAhnL0XWkY3y6Ihr0CG1W4YISQzUls14qIwfqTlwxJ7AY26AOAaNP+NP+e/ng2JO1Q24e3gVJb1pb6JMQmXnRKdTayrh4V48Ky75MDiTEQxSQJlUixOVAexCqU0qgox57l1iTvMIp+M+BtWKZEZwStQyNVR4LbsZHAMYehQdeeCkBDA1lVW13hQrt6hpBSV1pRRi5t5b74LkMU01m5QX77G0zc09jEilHd6FONpXp2K5EpMK/rYd/gAQDiZEFF8/wv2RFo5umKLkxwCOsimO00/PxzK3IsTQ4b8Q7lKOJK2pCJdSQ9tMU2KE2fRIqEJVNcMIx2at8j8SSq4SIYOSJYlFEm5b+pj90ti0tLjMqLYdEUnBx+rfAJdMsbaBle9hTZxqp1HxcyBcNkXcVEQIaZ6RYZkuRdBEALWLtaxqplnY3Jk6k4CPB48ES03noXwElJ8v7APtT6KfVX9w7cMu1CcesfzAq2ugXzWmF7JI5d8cB1MuHp6pRSw/HClxrP+KTAGJiGoT4YM7I2zuw8b9ft+2mYnWRVU5nciDy7oITOkUTLAqxI5Fm6qRtE2YW+tX0LBIjypgKGoT809tE3GmVDdHSZbhh5U8P6Y3xxdIoLGIFPAcae7h4UbhIqJM+zBKNFmaCIAxzN2EKczu72+td6aa0jgza2uI2MYUYWYpZ0A8AGeBFrhvZkAmFyXnOE3zso04KEFAFBpRBU0JjnFQssiDPSjKNpWYj76VKRgwd0s3YmYyVTTlx0oH2jfbA7dt3HfzhKjqaemxWwZEsok2YmZSQmMmhlvuNoUz2SG2nvB2337/+uPjtbXTaZ9IJIviQLSJVHBYy5X1lcz0QP5cIiIknMQENFCmh7dG2pkLJA2riWFtMqVUISHUFtlkFmHNQK2jBGs6BxOgAEXArBoZChSjQeVhTFnVSWSmGTLd3X0c+m9REYFIwt1LQn2wHmvIolTR19xrDSksYDiazgS88h7UPcML5bbz0n95Wc3exy6n5RLBczLa01/+7fN//7fX77eQzqxNT72b7mYuRI25ExNSQZ2ZhWfYRKpGUBJcT9imf/n2dlo/sZ6HMzhPTWoFMLKcGNkc5sEMSUQ66oVCMHMGhApUIaLkSG4gLfTUCKEsQprMHogA1yYkJxuhTYQ7N52wTBDEg44VQNqq9Kj3cUaShzYF2MkdxCxRpoueIBhgnuFBRHGgAOSZhggmx4NPdMi6BJEzbMxjlB+JoGOb9jFfKj07g2AqTEzXS395Of3x+4/7CJEui27Wvr3bf/3nP/7vf3277dCrpoiqyvnc96zeDKAsV4hAFKmmlgilUgrVkkbz3Pa5jXmNVsaJeCwWLa8bMzIvJDSYEul5SF/zaCZRWtLkSlrBUt69YKJCXSkBJ0JyEFFZ65tXBS9tqVB1VDgkRMxyAAceGRWy3QGO+t1JfojXM5DE0hblo+3N2tiQ7pVtsxiOfDSPBstAWW7Gwco9Zq35oKRkEf2a3u/3JPTW1/NZ+sLSp837DBhe7+O//Lf/8d/++19+vM0gZkhEKqlo726e04vEUrOLoxTj0u+VaUSm+Nxjd9zv+77vEStI8fDnO1Jf2k/KOIeoVGN/rDr423wW5IHtPjKIIdK6CBfxapqNfYBlOZ0zjgXWBRLMQrdBzKqtH9MQVmapeU+4B6JrK9JjRFWZ04dZzqrQ29q1dWZlpsw8bHtrWAlYWC2ROkCi+J/6rHIs/1taJypU7niO6e5zzsjoXUgXaWdZnwn7nv3t++3f/vL2T//tv//7X99ngBeaniOHZgSBhDnkwTJC8R6P9CXCM6xc2gCOoDFzu9u+72ae0HwAKNXSxDEaq26nWqjaLkyHz+IxKEUTWp+eCWTD37YbJxrL2tu6nK6XD/uc33/8YGXh1lvvS2/aiKkWI5g9piQBuEGq6s/qjmyOfcPDu8DTjzNT0eSsOJ4cRU3yauZrfxAwbSKosdJj3n7MKnEMN49HUfZ9NajMn6fvFobMaZhBaKf16ZfQ/fXd/uXfPv9//3//41/++rplLpdlML+PsWWoR/ix/O3gFD/+zghPKZ2MZ+2wiMgMhGPb/HbfpwWRZFh9mvokzHnYfKcgKY5XK3H434iICmsxK4UFtTWKUsCNmUDhNEZ4UGuLp3vktGBN1opFMYYRS+/d/DEhPGirRIym8DnSrKgeBxP+J5+NSFUB8kKXMzLSouhcHhFjDmImTcbRHwRDqs3C0ZPhmMccf224W6KYxyKSoOH2/W1cb+68DtY/3j//l3/54//45+/vG6hr64sNG9NToHgslqwigR6z4KgcFCwiEZheHsTBnGC4YbvPfcyapR5EFlBmEomyghpII2Aj3XFoLsvhvfaeVotz7GFhIaHMfZvfb7ftftu2zbx8+T0zl65PL9eXjx+vl6fe+9pP277/uL1VbohjLyZX9SlMXZpwC/dMlDwyi8eMmkj0zBxV1oQhEQgrrZ+bR4jC3Qtxo8Pn/bG6M3/ioPkYrxW1GFFzANGgeH2/v//ly801pW8j/+m//fs//evn7wPOJCz79PcxkrCuXbVrhvuYXpTKYx1xepQEnAobd48wV6A1BkUGtj22fc8HpElcwMrh9sWiCQ1PZTZzM6srXBclombwxMxj2/f7ZvuY2357397f3u/v72PbEyCBBxJoDU9Py8dPv3z88PFyfbpeLz9ef3z++q21hYlq1AVUZlBmfLhcz63ZtEQm06Ongoiw0NynIWaaOwxWcb742RHxoGEfVEUh1qYCeQhDDgC10kkNeaq3r77dkRF43+bnH6//8vv77uTQ/+tffv/LN+NVAd2D7tuY4a3r0hcVVRcvSUqCDkZVZX93JgoyYWHwMBOltS9oY2w+LMYcR3ote+oqyRisrIeGghaVKbETezgzU3IaPN3MM9ISP759//LHl+9fvm3v+9wRASU0weWil0t3nxZunrcf+/3Hv/1F/kN660uv97D1tTXNhJlNs8ziwcn383kRcZsQ0abclJhFtZ+Wxfs+piGkK4smISKtSG5gFlqWzky1Z45AKtJUCZzz4AFV2DnsGZPycBojZgaTuTtoJv94n398+/HlR/JJv7/HPaTxwm2J4c7B0qTJNNOiIyVg4aW0iNpUmekeQSB2UWVQOiDZdCFmGze3MKv88dANEA5/EAYzZRBxttaJ3CPCDhgrjsY4X79//6//9N9sTnJatH348EwBH2bTEGF7fL3fTiesp/WyaBZHj7n35XQ6Z4RHrOdzaz0z932737dhToBAcsa2z33bZnggLWxM8wxpKl1++fXXy8tzJ3RRql1f6QBYRZVb74Ra5F7PhJnLA+uxHfcA4n6OI4+UzMKpaghwb8sZ0l/vt//4gtNL6Lpm+Os2XtYnXrLxoQjYtndtyxIRfV0jYm77bVgTVpGkZCFmbr2BsJ76elpq5e/YfY7khshg1tY4aZs+5Tj3GNueSstyWZdTBGnDis4Ec6uJWHjYGPf7tu+71CLgCDPv3Ii48PcuzMrk7nfjBhEl1HYbs9xFBR6vX3401dZaZkpQg2bE3Hd7vyOP4qLATgYlkXv4Fn/5j7/Ql89t7ev5vF7O5/NpWZfeFUyEZGZhItT0DMdmay8OSmUFMIsSBYXVrUAQkWfMMU7ny+t9vN722x4u8vSBXNtI1XWlVbbwsGBhUYGj967EzKqiAiKL8Hkw94TlEM4tDUnH6MHT4apyvrQmHoH32/Z0ubKom+FgvMLMMobq0jQzgqjoORxx8JkIGR69tf/lH//Rh739eL2/34aH9NpHkkykTAoPS7hlUgE50+bw7c7vy7owy76PXNb2xL313nvNGqfNt7fXfd+S0uPh+kf4Cfl+/zFScHlepLUzobe2nlZtrYgmlQIEkCLmlrt5HlwKHK1H7UVnyqyQwaCJHDbBOj3uu+0GkPaT7NlmCLW1q97vd0cqQ5gITNrUM2po5igGN6K4eiJFLxUWYp5j7PuWFqd1OZ9P55MiNrf8/v3Hr7+cT70hjRBFJisy/dgHUkSEWFHiTeGD/yukyk+Xy2+fPv349mpzzH3v3C7XkzIr9Ol8xhzffv8rK5hIuqooInebhYMrCZM6TAAlWbQzKwFmc1rhrOIZEdPMI9xqqSYhGaJ0uq4fP3385bdfri/XZV1ZlfgRxlG2qzVfPXSX+DmOqSj6c5L/k/pLhKDpvvnYpw9PC4pUd8axsZAJUGGkEx0gK4jU0pmJm0pvPKe51brBBzsLHl6Wd0TBSqosyghxJ3Pbxz4tFmixPQ9tQ01NI9wNAB85ikT4WNDO1JqiUWu6rO16PQvQpV3O5y56PV3/8e//U47xfxU5oCAhEQZ5eBGNiMXMCcU9ZRVh4jnn/Xb//OVLUmrXiuCq6n4MvCJhnn/35w+//Pbp6cPL88dnXdtBOHSoKgtzgoj5ICfiwEbxt/Hl8RDw6IgB5EFUscjd4j5tTDNPj/KXLUgRoGQK4Ti+U4KIdJqpiDRd1tOclh7VzTqSD0769DmJ0Jt2aa1phI/t7nPT1ohketgMqvYLRKUPYRFWSqqOmkVYKJOrBCWm1jgj3a13+fjyfGp9bPv9/fXr7f6t9fTx68uH3377ddy32/0+5hj7qO98gEoe4bksy/l6eXl5brq4h7v3dT1dLl9/fH3fb3mwvi2QotKbsGpK/sM//Kdf//SLLk1XzcSwGeEsrExdpLzL//ay58Emigc9Iqs7OPiyxCX8yExQCmzkmD7ssDut/YTgLOYdwQFDcf6TEKljzsgUIunaekv3tPKYr1lhRLhPb0rCKYKEzRn7PsPhHlkYbpAc/XoRD4RIj+HXMVyNQ59VnLsAmCPCp7Uujdel0U3y23b78eNtu+PH17c/f7pc+mXu+5jDppFIX5alL73XogqQyPV6+fjpU+vt8x9f//3f/+P17bWKushk4n1s231MTxKcr/zhw4enDy/rZX35+GFZOskx8mKiptyaruuiTe/3ez0DqkFBqX7D6wEcfcBR/BzUgwpKAQbJjDnM5/SwpAAiQF4blcKcSn0epd8VTqiZZUSoJCCirbWA8xHVEoXCdRZON9ssmElZawN9RGzb3eaMVOaHE2alqcyMtPCy1MxkFYjqIT+opMiZSOEkhZ772vW8dBX+/Pu37X28vufT9SrUVSgY08L2fZuDbhXGIKxvt/e7jWVdX3+8vm7vt7FVJJ9mtfMTnL3peurPH68ff/n08uHD6bKyMjKKT0sMFiYWUabaRF5ck0z48e5XyV/wItHPXgw4EJgEEF7b5Gl67tPGMLOy60qC156lxBAKUBh5ka0TrHVXIpMyWVhbS5IShxadloJUnNJthM0Qha7cuhSGdL/fx5jTOzEEh9imgOfMcKQe9rIOQu0TyoTDI5NBvakwwhOZbe3X82lZZG3++a9fZ8Tn729VDDqXovUYqwdSRAH+9v7j3fbz6cwi6/VMXfZ9v72/79skYD318+V8vlwuz9fT9bSe1t77Y8JFAB0i8bq64XP3wN/IJigJd8ki8whKREyHC0hWkRpUpIvwhAXtc27b2Mcws/DiW4EDSZ6YLCQgSjKugxL1PKBAZmYVIU5xTC8eA1MmR4kPii5EBDz6rGKh7/vct0HJ3JgzkKQHYf5gJYRbZBBnpjCICV5EfSJV6U190rDNfDRdL9eT//JJRL5+vX358XY6S2sNQApEdVnXdV1F5LSeSeT1/Y1Jlr6QkI7JW6O77DauSzufTqfT6XK9Xq7X0+WkXR4SJaegY+1bAajV4x0CFdDP3edHrMmjtqiIU1rWSgJ0/JsaG3hgztzH2PZtH2PaoRyDKAGM0j8fcToLJAErHeDSMdFhRpY4Mmemg4IyfdaahWpxGUfZEMJEqvuw+2ZMTZBSC1gV5RuA6tXr4wVHmAg94ERExB57a8oqrJIeSbSeTu1P7fnTpz/ttm0hupj72/vbj9fXMaZ57Ptg5g28ns79+vz89LQsy/v9fvvxHdmfL5eX335pIl2kjGCkMSSylMNCjdsYe0wT6hk1as8gHHN/kfuYdDgsPF77R7A5ul4+vCLqCkWQg5zIE3Ni7jnmHFbDHQuQIAiNHqvmiBHKxfchkPYmZRMQHiS99WVdL2Hj9vr9Pjcg1tYAIitEnDJjjEGACENlGP1428/nc3s5sdK4v1OMfqa2aKSP/eaTytXcwva5uc+ahDBLTSr3OSip9iRu5sMMSQExlVgxnUaGr+tpWU+Qori4xx45A01k7lMsMh3LKrU4l1JEtEtr3GvOQ0fZVMBlxEgGJxGaiFbBUcZNkeRF86vbfcyBUItT6iIUE3BmIoIgUX0Nk4e/3W63+7H9s9KkRUZaoWVZrgssazstSmZhM7SWcGUgLCymkaQ0IpCAuDY8m9QCeWIB1V1LKvd49sTr+3351pfWny5aVKdpc99rfYSRKIPzkZejmvY8XImY5QgAdScOcWAm4BAjGogd6aS1ug3EYTFh2qX1DmCHwyaBRFSUF22qrOTCwa2sb6qImWUIUcFQhIs0WkOYqmOKmxIlms1HzX+A0EmgQFmUElrjGlImgSgiHbTN+fa23e5zH3VECPqpljMmVWlgKa9WECAgQI91zhY23DFzBmX2xqLSendLoIwTqBTuEYTaxRkZnknYp337/tqEGNeny9KEA/M+dmGwCos8lH/0cDbPR2fPQlKVU/w8gDx+IACQgrPS3vHnSn3FaEs/nU5zTvNMd+IsUe5pOS1dECNip3J2Cwu38JnpNeXuvalqFTPhZqBAzYuYOd1rLgLgaNvxGGcn0i1Yu4iEZ5iDKJkDOWPcNvv+en/b5j7TArXw/XgATFFjuAjM6ceEKoNS5zjQ4TEsp7sMAvRyatp59TkAN04XIhEGcc1pEJkRTs7OGbHt+9vt/XJup9Oy9E7gWmNJdFgmH0ObY5Zazj0lo0oqDnj+ZLqCcNw4JQFYQI/l0hqgkBBHEyIEIxoByk15abp0VUmuNzLj0FCHhU1EmduDuJxdOUuplsgSz/wkhh/nXiX+I9U+GFy1GjLB5rbPIC5EJMbM23283sc2YgS8KiR65GppkFb7A71W1rojkQT1aTUyQqa7DfOhe6699a595UzLlEw5ZpYMBnMJyZFle4q0iH3M1/ebNgFO5/ICyTnDqAaBVYAWI/YxlkFmqTspA+k4tkMXmMREogIRzlaPTQKCTIuc7pHhYxMQM0RlaWUMj/SxD4+cXNZXyPRa+JDEtU+WgUMtjr8RczOS6npFgQYPiAcP3UCUjTxnBFnGmGEOhjDU3N43e7vP4bBgJ3I+6KIQElYwa1t6X83MzNSKrJSRqUJ0bGxLIaslJQeRkCFlblVuU+UuRceGJa6+PNynwT2Jdo9537bx6eVPv3269DU8bVhrJD9xrAeefqg3jv9/AIxgQuZDp1wMSSgX6YGJFeDSqCinWZGxWYVak95EhDI83HzugJdty8PHSZCHFwrj4G4Qy+GvBEIW0ZYSJL2h1sf97UYcQ5cyaIgMT0lqrKS6BtE28tuP7evbPlMmyNGqquUSJxIzK0gO7ZIySIidLDxCl9YjwJCgIEuDc3LMjBksWaJhLp7uYwkzgUR+6kAwdhjDI7fN399vnljO53Y6i6yQECEuliMevBU89KKAslS8ETn8PR71NxEg8DL34TIsA4wAEIMba72bQhBmQUo6EMkQFYtIImGSSiBJR4g7yB4FJlMRXpIQdCA+XjxDlgPrOXC3siVLz5iW4ASLtCbSQPK+ja+v979+/f7t9d2ilUgmyMv3hB8OczanzWLulVnRwWtRFc36p8KqDcEIuIWbK5NI475wzPQZHh4l9VVhIaJpHpa1sjh27Imhye399PWHLOfnp3PvZ2AiE2n/k+4kMg7ScqlUmFFUqMetiEziJDlmA0FJJbxmJDM1ZhT4VD9QTHMk1f7R1qfD0+SAR+h4heMnuwc/iUx/a97rlvPRZtVnKcwhqiP2Aio5anuWqCe/37Y/vv741798/uPL6+02pkqxDw7d0eHEeGTCg8FbnHpQSHqEpieS3MNm5CH/yNo5ztR0IV6abW/DhoWbJTGaKgtTsgccoc3nxDQACMLrDf/x+/eUVfp6ej75boXq4VDhHGdA5eNeYDj//KAPDDJSkjUVUfPp41YIi6qKNgs3M9XDhqIKmiwcXYhIPEoLdpgW0dE2HbJ9IgrieJCED3qXMjM544Fn4VFthFnWViZpPZMDsOnv2/7H5x//8u9/+ctf/3i7jUg2CyJpLNyAYKEkLifCZISw1himGgojlELGkPCZc4Ybwp2I3bPWeTJr7z1srwgY6QgoUBrKQgGLj+IotRftM79837l9v1xfLqeFzRri8aUeb+AR9qlWOh31KZcxYg2YwEE1k4pjAz3wkPE0Zd+mz6HchZoeXkgly0xEsVGrynjAZ8yFI9Ah9D2AEs+MQ8tT7BOdDxEAHZcxqi+uCnntPSznxH3a1+9v//qX3//j9y/ffrxZcorCkyVBkLKuRm1yqqbCYtjD34sjq0eDhmcmipkUlshMhnvcbluEua+JRiTal6NPAWXS9LAorW8IMwsWzlqnXlafP277f/zxRXL8+UXPp8bEtdhLREA554xav1501OTMeuRRrZqweITtm1SoATELiDJ9Ws2WU5nDfHqWOzzXvpbSWaRbeDpUpC9NWNzmGLtHMDCmlegAfLjbHUtEHhsFDYcEno9VTERaGIpkUlhxpvD2/v7P//yvkOaQ922SpLYecx5jCySoGN0sLBGYNgNkh4cDg0VZtBg74SiRCA6DT4qEe06LOY0JYE2aQRKZI5IP/5Bj+QZVmCYOIndMg+cU+dZoPq8vJxXljGOL5ENRCFTgKTZVGbwxOCovEJMwtTKVYZCAtc6rIuwcw82IOTJAxEU2fBSXj7khkjlLHZyBevuISYSQYE6qwXsecjYWYonw4uHSY6HL0QEgw91834bvzvct3m/bNoYskiCuRomZyngmkxAMogfDk4maikcSSdmVEiux6L7vkYiQcEoLHIZWUXoEMAJ86gyRFE0qvbIzUoiJpYmUWo2TiCTJI7EN5O7uN4n552c5S1+7EHN5LdZjIGYiOSTK6f//ps5muZUrh8EAyW75ZpK8/3ve8djqPiRmgdNONtq4SnbJLf4CH2eYSTzeMg1CyMjAbmn0RGaJFjzPXoIapvscNt81O8YIv2k0Q7FtL2A72oBiDNh43somiwiNBafE/p/BaoTpuXvW6vfS1fz8un5//vfqrhkxMs9wSW/u6mgHUoiaaVXWcZ5rMBNLGhFRiKr3da+WkJw0w/hI7CDVuvt6X+RffxxHDAsxolqrpydwZFWm0NzUHHu2l6B78PXW78/r8/Pz+jXn8SvwD91CkgdE675xi2RkHJVH5jMd6lJmlDnPgoat1qbvPEHJ7kQb5US/BoP9oEwkqTtce24bVzwGVLS0BGs15Fw5G40OL0xMYnR3sfpuQ9IJ5H2///f1NbOdKTCJ2qs+ar8AvsCgH8ngnncQCLEUVQ01No93uSLjhKhGT9+j+8ZxHB84pYjjlYwRe9YI7SpyH2oCHp1nHkLKfdVaa3bhts9iP030s2n1ZmNkMxpseVwehXNPJgA3IYOGMEDVceYxWvOTHw0y2dtaKhyLwjWvIZQgxWw/0FALy1R1/NiztZZvem0sSkxP97rXvXr1CBlxFHKk+76scRVMu5bhxj864eCDgQMVeXcI0WbSsRSlqDrPI4bToUWqKTKLVgdIwiz1972QWWVZc2kSentgtq/Vc5/VcWzPjL2RiX70EBmZS2tvI4VpUcosd3O7DEIF0+8RinnAhvPkwqCPGSIyGUkFxh0G/onmrswi9e+Z5gZW+OdosaW2dNoQPlvkW+2/DU4qJpn2rCXrANFxVFbM9H2/WVDYSRa2/gcSYIYPSbizDPo3LiA4LLGax/AEqz7++HNG68Z1STEcVkYeWbsRaTLumVideTBPpg4EhLWu0aBVVf6Ylr2D+/uM1nR7GGBezm58noGoRkrEXjMioZwReveI0/q+Lg/iTXEMkkxEWSm+POtxD+XDApIfujbR3EbZ5gixaXicQYM29xpUCyvlf4yD7XYRdPvyyKAtj7juOXISWOv+fl8OgP4GCGyLDcDdTgcFQyYxg3sAJFgdR/MYHsJRx+ucocHw4EIjMyszU27fAFd1aoYf7hBtFLfpWvMEH5lbQTKJWY172WKz478jEEmDtKY1agfmkda0KN4WB0b33O87wsWJIsSqyAktKFxKbbMxnubK01/SkTyZVmS4FPemsbt9aHZkf7EUMUZ89kaISNQszmAm8CwqDfUBKhCJ7nm/XTG2nNIFoYalICIVOSEipyHQmH6xJl4Tx/BoHg3rfiWXhNFp3A0iRO0wuGvkLVji1u2lItQjaG21zB7eMpiEBpjVwo+xQe7oKYZ1rgaXK21cdV+tAZSRVZSM4Nz03iHYQ7WXA35KPelKMveI1fWRtDiS8awjyr2FYRU9dqy1tCVowog9c3X30ln1pGkXn9umGkRGxMfx+vjQcQhxXehAU8v6KwaiFKUMVSpiuLOWT3rm8WoW4gQPsQbZYs1ygbaF+xu4wWAE9t7qKQg2qfVAJs+TMxJXr2WJuhWglRI0HdKRPNPuee509Ew6vI/sAZiDlAye9SKAYjqxq8oP177r4Wr+Z4wDhCIEBQfoeFI294bWc8NBNoOyuVf32HOww455ENLc00s9mmAVn62vBChCm2opTuT58XGh/LFF1d0asTsRGVWKmoyOnAg4K1d4gYw8O3J/S5CmzpVmPLESmmhRYFotY0JOBPvd0/d9ybcZ6jzP/CXWrc/1jcaCVMHXcWTF9Lq/b2FeL/z9wlx3ZR1V0ruKEeq+85XB0t3Io4ezXJ7DwU/BfsaX6fIJULdHZ89GIRgQYfGYxnK2vdYy443TZDITyB50a0mjakxj10mCD7nP+E5FokInSebC3EvAZEZGiHE1vq4rMu93f7/vKHz8+Z/+mtW8l3r4K4sZw7yUahDBPM/j9To/Vuv7vaTamzWNLwX8H7EHaKXoRa+SAAAAAElFTkSuQmCC\n" + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ], + "source": [ + "# Check some sample images (Use of cv2)\n", + "import random\n", + "import cv2\n", + "import glob\n", + "from google.colab.patches import cv2_imshow\n", + "\n", + "TEST_IMAGE_PATHS = list(glob.glob(path+'*/*.png'))\n", + "\n", + "images = random.sample(TEST_IMAGE_PATHS, k=5)\n", + "\n", + "for image_path in images:\n", + " # print(image_path)\n", + " img = cv2.imread(image_path,1)\n", + " cv2_imshow(img)\n", + "\n", + "# Your code" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "jWBEMC1FUfXS", + "outputId": "81aadca0-a4e1-4af9-a6e9-ed0a5565ad0b" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "(128, 128, 3)\n", + "(128, 128, 3)\n", + "(128, 128, 3)\n", + "(128, 128, 3)\n", + "(128, 128, 3)\n" + ] + } + ], + "source": [ + "# Check shape of the images in your dataset. This will be helpful while specifying input_shape in your Transfer Learning Model\n", + "for image_path in images:\n", + " img = cv2.imread(image_path,1)\n", + " print(img.shape)" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": { + "id": "52BhBWRab5yc" + }, + "outputs": [], + "source": [ + "# Check if all the images have same shape, else you need to resize them to some common size\n", + "image_count = len(list(data_dir.glob('*/*.png')))\n", + "TEST_IMAGE_PATHS = list(glob.glob(path+'*/*.png'))\n", + "count=0\n", + "for image_path in TEST_IMAGE_PATHS:\n", + " img = cv2.imread(image_path,1)\n", + " if(img.shape!=(128,128,3)):\n", + " count = count + 1\n", + " print(\"All pictures are not of same shape\")\n", + " break" + ] + }, + { + "cell_type": "code", + "source": [ + "# If the shape is variable, reshape to a common size \n", + "if(count>0):\n", + " print(\"shape not same\")\n", + "else:\n", + " print(\"shape is same\")\n", + " SHAPE = (128,128,3)\n", + "# If it is same, prove it" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "CqPABIJi_MsF", + "outputId": "e5251b04-a9a3-43b5-9773-4c7bdd881b2b" + }, + "execution_count": 17, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "shape is same\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "zSoUXS1cRbnu" + }, + "source": [ + "### Model Definition\n", + "Choose a model for Transfer Learning (You may also experment with multiple models and keep all of them in this notebook)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "QKZmIgXMTHfy" + }, + "outputs": [], + "source": [ + "from tensorflow.keras.layers import Input, Lambda, Dense, Flatten, Dropout\n", + "from tensorflow.keras.models import Model\n", + "from tensorflow.keras.preprocessing import image\n", + "from tensorflow.keras.models import Sequential" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "9xWLUibHRNGj", + "outputId": "50424f75-164c-4279-e057-7790279f0baf" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/vgg16/vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5\n", + "58892288/58889256 [==============================] - 0s 0us/step\n", + "58900480/58889256 [==============================] - 0s 0us/step\n" + ] + } + ], + "source": [ + "# Choose and define base model\n", + "from tensorflow.keras.applications.vgg16 import VGG16\n", + "\n", + "base_model = VGG16(weights = 'imagenet',include_top=False,input_shape=SHAPE)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "J3TwB_GLd7BU", + "outputId": "cac93d47-d910-4819-eba0-057c91d31cab" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Model: \"vgg16\"\n", + "_________________________________________________________________\n", + " Layer (type) Output Shape Param # \n", + "=================================================================\n", + " input_1 (InputLayer) [(None, 128, 128, 3)] 0 \n", + " \n", + " block1_conv1 (Conv2D) (None, 128, 128, 64) 1792 \n", + " \n", + " block1_conv2 (Conv2D) (None, 128, 128, 64) 36928 \n", + " \n", + " block1_pool (MaxPooling2D) (None, 64, 64, 64) 0 \n", + " \n", + " block2_conv1 (Conv2D) (None, 64, 64, 128) 73856 \n", + " \n", + " block2_conv2 (Conv2D) (None, 64, 64, 128) 147584 \n", + " \n", + " block2_pool (MaxPooling2D) (None, 32, 32, 128) 0 \n", + " \n", + " block3_conv1 (Conv2D) (None, 32, 32, 256) 295168 \n", + " \n", + " block3_conv2 (Conv2D) (None, 32, 32, 256) 590080 \n", + " \n", + " block3_conv3 (Conv2D) (None, 32, 32, 256) 590080 \n", + " \n", + " block3_pool (MaxPooling2D) (None, 16, 16, 256) 0 \n", + " \n", + " block4_conv1 (Conv2D) (None, 16, 16, 512) 1180160 \n", + " \n", + " block4_conv2 (Conv2D) (None, 16, 16, 512) 2359808 \n", + " \n", + " block4_conv3 (Conv2D) (None, 16, 16, 512) 2359808 \n", + " \n", + " block4_pool (MaxPooling2D) (None, 8, 8, 512) 0 \n", + " \n", + " block5_conv1 (Conv2D) (None, 8, 8, 512) 2359808 \n", + " \n", + " block5_conv2 (Conv2D) (None, 8, 8, 512) 2359808 \n", + " \n", + " block5_conv3 (Conv2D) (None, 8, 8, 512) 2359808 \n", + " \n", + " block5_pool (MaxPooling2D) (None, 4, 4, 512) 0 \n", + " \n", + "=================================================================\n", + "Total params: 14,714,688\n", + "Trainable params: 14,714,688\n", + "Non-trainable params: 0\n", + "_________________________________________________________________\n" + ] + } + ], + "source": [ + "# Print base model summary and have a look at the layers\n", + "base_model.summary()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "F_Heq3C1eKd-" + }, + "outputs": [], + "source": [ + "# As we're using Transfer Learning, you do not need to train all the layers. Freeze all of the layers or train some layers (experiment)\n", + "base_model.trainable = False" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "MKx1EtUJea6D" + }, + "outputs": [], + "source": [ + "# Append Fully connected/custom Conv2D/Dropout/MaxPooling layers to the base model\n", + "inputs = Input(shape = SHAPE)\n", + "x = base_model(inputs, training = False)\n", + "x = Flatten()(x)\n", + "x = Dense(2048,activation = 'relu')(x)\n", + "x = Dropout(0.25,seed=1)(x)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "q6I3oTTNgP8L" + }, + "outputs": [], + "source": [ + "# Add the final output layer\n", + "output_ = Dense(3,activation = 'softmax')(x)\n", + "model = Model(inputs,output_)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "6aVQocJwgN5r", + "outputId": "dfd31a18-0811-44f3-d0a6-d2806b20e9cc" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Model: \"model\"\n", + "_________________________________________________________________\n", + " Layer (type) Output Shape Param # \n", + "=================================================================\n", + " input_2 (InputLayer) [(None, 128, 128, 3)] 0 \n", + " \n", + " vgg16 (Functional) (None, 4, 4, 512) 14714688 \n", + " \n", + " flatten (Flatten) (None, 8192) 0 \n", + " \n", + " dense (Dense) (None, 2048) 16779264 \n", + " \n", + " dropout (Dropout) (None, 2048) 0 \n", + " \n", + " dense_1 (Dense) (None, 3) 6147 \n", + " \n", + "=================================================================\n", + "Total params: 31,500,099\n", + "Trainable params: 16,785,411\n", + "Non-trainable params: 14,714,688\n", + "_________________________________________________________________\n" + ] + } + ], + "source": [ + "# Print your model's summary\n", + "model.summary()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "qdC71fUBgXAg" + }, + "outputs": [], + "source": [ + "# Compile you model (set the parameters like loss/optimizers/metrics)\n", + "model.compile(optimizer=tf.keras.optimizers.Adam(0.001),\n", + " loss=tf.keras.losses.CategoricalCrossentropy(), \n", + " metrics='accuracy')" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "RdUSMLggifex" + }, + "source": [ + "### Data Augmentation and Pre-processing\n", + "Augment the data. You may also try dyanamic augmentation using [`tf.keras.preprocessing.image.ImageDataGenerator `](https://www.tensorflow.org/api_docs/python/tf/keras/preprocessing/image/ImageDataGenerator). \n", + "You may use [`tf.keras.applications.vgg16.preprocess_input`](https://www.tensorflow.org/api_docs/python/tf/keras/applications/vgg16/preprocess_input)(or some other base model's utility) for pre-processing (can also be passed as a parameter to `ImageDataGenerator`)" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": { + "id": "DBscSsvkgn39" + }, + "outputs": [], + "source": [ + "from keras.applications.vgg16 import preprocess_input # Change according to your base model\n", + "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n", + "\n", + "# Your code \n", + "datagen = ImageDataGenerator(\n", + " rescale = 1./255,\n", + " rotation_range=20,\n", + " width_shift_range=0.2,\n", + " height_shift_range=0.2,\n", + " horizontal_flip=True,\n", + " shear_range = 0.2,\n", + " zoom_range = 0.2,\n", + " validation_split = 0.1\n", + ")\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "IcKPxCpOkcuG" + }, + "source": [ + "### Training and Validation Dataset \n", + "Split the dataset into training and validation (We'll be looking for your validation accuracy, assume we are using complete dataset for now). \n", + "\n", + "Hint: `flow_from_directory` used with `ImageDataGenerator` will simplify things for you." + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "sB7hb3ybkJRq", + "outputId": "0acbc10a-e1ed-43e0-f7b5-a6ea77fe643e" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Found 8094 images belonging to 3 classes.\n", + "Found 898 images belonging to 3 classes.\n" + ] + } + ], + "source": [ + "# Your code\n", + "train_generator = datagen.flow_from_directory(\n", + " path,\n", + " target_size=SHAPE[:-1],\n", + " subset='training'\n", + ")\n", + "\n", + "val_generator = datagen.flow_from_directory(\n", + " path,\n", + " target_size=SHAPE[:-1],\n", + " subset='validation'\n", + ")\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "ZZPsjpT1mp3z" + }, + "source": [ + "### Training \n", + "Train your model for some epochs and plot the graph. Try and save your best model. Experiment with the parameters of `model.fit`" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "Gs2X14MBmu7W", + "outputId": "bd9b3bf6-8c5b-446a-a661-be11952113a1" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Epoch 1/20\n", + "253/253 [==============================] - ETA: 0s - loss: 0.5189 - accuracy: 0.8614INFO:tensorflow:Assets written to: /content/drive/MyDrive/Mask_models/vgg16/assets\n", + "253/253 [==============================] - 103s 354ms/step - loss: 0.5189 - accuracy: 0.8614 - val_loss: 0.2201 - val_accuracy: 0.9143\n", + "Epoch 2/20\n", + "253/253 [==============================] - 71s 282ms/step - loss: 0.2115 - accuracy: 0.9192 - val_loss: 0.2449 - val_accuracy: 0.8964\n", + "Epoch 3/20\n", + "253/253 [==============================] - ETA: 0s - loss: 0.1767 - accuracy: 0.9369INFO:tensorflow:Assets written to: /content/drive/MyDrive/Mask_models/vgg16/assets\n", + "253/253 [==============================] - 78s 305ms/step - loss: 0.1767 - accuracy: 0.9369 - val_loss: 0.1778 - val_accuracy: 0.9376\n", + "Epoch 4/20\n", + "253/253 [==============================] - ETA: 0s - loss: 0.1713 - accuracy: 0.9371INFO:tensorflow:Assets written to: /content/drive/MyDrive/Mask_models/vgg16/assets\n", + "253/253 [==============================] - 80s 317ms/step - loss: 0.1713 - accuracy: 0.9371 - val_loss: 0.1518 - val_accuracy: 0.9388\n", + "Epoch 5/20\n", + "253/253 [==============================] - ETA: 0s - loss: 0.1538 - accuracy: 0.9453INFO:tensorflow:Assets written to: /content/drive/MyDrive/Mask_models/vgg16/assets\n", + "253/253 [==============================] - 83s 328ms/step - loss: 0.1538 - accuracy: 0.9453 - val_loss: 0.1332 - val_accuracy: 0.9510\n", + "Epoch 6/20\n", + "253/253 [==============================] - 77s 304ms/step - loss: 0.1534 - accuracy: 0.9433 - val_loss: 0.1494 - val_accuracy: 0.9499\n", + "Epoch 7/20\n", + "253/253 [==============================] - 75s 296ms/step - loss: 0.1276 - accuracy: 0.9524 - val_loss: 0.1370 - val_accuracy: 0.9510\n", + "Epoch 8/20\n", + "253/253 [==============================] - 76s 301ms/step - loss: 0.1390 - accuracy: 0.9486 - val_loss: 0.1402 - val_accuracy: 0.9499\n", + "Epoch 9/20\n", + "253/253 [==============================] - 73s 290ms/step - loss: 0.1329 - accuracy: 0.9522 - val_loss: 0.2010 - val_accuracy: 0.9287\n", + "Epoch 10/20\n", + "253/253 [==============================] - ETA: 0s - loss: 0.1214 - accuracy: 0.9558INFO:tensorflow:Assets written to: /content/drive/MyDrive/Mask_models/vgg16/assets\n", + "253/253 [==============================] - 77s 303ms/step - loss: 0.1214 - accuracy: 0.9558 - val_loss: 0.1175 - val_accuracy: 0.9610\n", + "Epoch 11/20\n", + "253/253 [==============================] - 71s 282ms/step - loss: 0.1130 - accuracy: 0.9569 - val_loss: 0.1114 - val_accuracy: 0.9577\n", + "Epoch 12/20\n", + "253/253 [==============================] - 72s 286ms/step - loss: 0.1179 - accuracy: 0.9576 - val_loss: 0.1202 - val_accuracy: 0.9555\n", + "Epoch 13/20\n", + "253/253 [==============================] - ETA: 0s - loss: 0.1025 - accuracy: 0.9628INFO:tensorflow:Assets written to: /content/drive/MyDrive/Mask_models/vgg16/assets\n", + "253/253 [==============================] - 76s 302ms/step - loss: 0.1025 - accuracy: 0.9628 - val_loss: 0.1077 - val_accuracy: 0.9644\n", + "Epoch 14/20\n", + "253/253 [==============================] - 74s 292ms/step - loss: 0.1098 - accuracy: 0.9603 - val_loss: 0.1126 - val_accuracy: 0.9644\n", + "Epoch 15/20\n", + "253/253 [==============================] - ETA: 0s - loss: 0.0999 - accuracy: 0.9657INFO:tensorflow:Assets written to: /content/drive/MyDrive/Mask_models/vgg16/assets\n", + "253/253 [==============================] - 78s 308ms/step - loss: 0.0999 - accuracy: 0.9657 - val_loss: 0.0908 - val_accuracy: 0.9655\n", + "Epoch 16/20\n", + "253/253 [==============================] - 75s 298ms/step - loss: 0.1120 - accuracy: 0.9595 - val_loss: 0.1015 - val_accuracy: 0.9610\n", + "Epoch 17/20\n", + "253/253 [==============================] - 77s 302ms/step - loss: 0.1019 - accuracy: 0.9640 - val_loss: 0.1197 - val_accuracy: 0.9599\n", + "Epoch 18/20\n", + "253/253 [==============================] - ETA: 0s - loss: 0.1000 - accuracy: 0.9645INFO:tensorflow:Assets written to: /content/drive/MyDrive/Mask_models/vgg16/assets\n", + "253/253 [==============================] - 77s 306ms/step - loss: 0.1000 - accuracy: 0.9645 - val_loss: 0.0971 - val_accuracy: 0.9688\n", + "Epoch 19/20\n", + "253/253 [==============================] - 72s 284ms/step - loss: 0.1020 - accuracy: 0.9638 - val_loss: 0.0918 - val_accuracy: 0.9677\n", + "Epoch 20/20\n", + "253/253 [==============================] - 70s 277ms/step - loss: 0.1080 - accuracy: 0.9597 - val_loss: 0.0934 - val_accuracy: 0.9655\n" + ] + } + ], + "source": [ + "from gc import callbacks\n", + "from keras.callbacks import ModelCheckpoint\n", + "# ModelCheckpoint is helpful to save the model giving best results (brownie points)\n", + "\n", + "ckptpath = '/content/drive/MyDrive/Mask_models/vgg16'\n", + "model_checkpoint_callback = ModelCheckpoint(\n", + " filepath=ckptpath,\n", + " monitor='val_accuracy',\n", + " mode='auto',\n", + " save_best_only=True)\n", + "\n", + "history = model.fit(train_generator,\n", + " epochs=20,\n", + " validation_data=val_generator,\n", + " verbose=1,\n", + " callbacks=[model_checkpoint_callback]\n", + " )" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "FTvRa1FXri4R" + }, + "source": [ + "### Evaluate the performance" + ] + }, + { + "cell_type": "code", + "source": [ + "# Fine tuning the model\n", + "\n", + "base_model.trainable = True\n", + "\n", + "model.compile(optimizer=tf.keras.optimizers.Adam(1e-5),\n", + " loss=tf.keras.losses.CategoricalCrossentropy(), \n", + " metrics='accuracy')\n", + "\n", + "history2 = model.fit(train_generator,\n", + " epochs=5,\n", + " validation_data=val_generator,\n", + " verbose=1,\n", + " callbacks=[model_checkpoint_callback]\n", + " )" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "49_Xhnf6S9YF", + "outputId": "1a23bbbe-9365-41e0-db7f-ab36e886514a" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Epoch 1/5\n", + "253/253 [==============================] - ETA: 0s - loss: 0.0844 - accuracy: 0.9701INFO:tensorflow:Assets written to: /content/drive/MyDrive/Mask_models/vgg16/assets\n", + "253/253 [==============================] - 97s 367ms/step - loss: 0.0844 - accuracy: 0.9701 - val_loss: 0.0631 - val_accuracy: 0.9822\n", + "Epoch 2/5\n", + "253/253 [==============================] - 85s 337ms/step - loss: 0.0470 - accuracy: 0.9838 - val_loss: 0.0765 - val_accuracy: 0.9744\n", + "Epoch 3/5\n", + "253/253 [==============================] - 86s 340ms/step - loss: 0.0337 - accuracy: 0.9896 - val_loss: 0.0727 - val_accuracy: 0.9755\n", + "Epoch 4/5\n", + "253/253 [==============================] - 85s 335ms/step - loss: 0.0377 - accuracy: 0.9868 - val_loss: 0.0696 - val_accuracy: 0.9788\n", + "Epoch 5/5\n", + "253/253 [==============================] - ETA: 0s - loss: 0.0307 - accuracy: 0.9899INFO:tensorflow:Assets written to: /content/drive/MyDrive/Mask_models/vgg16/assets\n", + "253/253 [==============================] - 90s 355ms/step - loss: 0.0307 - accuracy: 0.9899 - val_loss: 0.0414 - val_accuracy: 0.9855\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Plot training & validation loss/accuracy values\n", + "import matplotlib.pyplot as plt\n", + "\n", + "fig,ax = plt.subplots(ncols=2, figsize=(18,5))\n", + "\n", + "ax[0].plot(range(1,26),\n", + " history.history['loss']+history2.history['loss'], \n", + " label='training loss',\n", + " color=\"brown\", \n", + " marker=\"o\")\n", + "ax[0].plot(range(1,26),\n", + " history.history['val_loss']+history2.history['val_loss'], \n", + " label='validation loss',\n", + " color=\"green\", \n", + " marker=\"o\")\n", + "\n", + "ax[0].set_xlabel(\"epochs\", fontsize = 14)\n", + "\n", + "ax[0].set_ylabel(\"loss\",\n", + " color=\"red\",\n", + " fontsize=14)\n", + "ax[0].legend(loc='best')\n", + "ax[0].grid(True)\n", + "\n", + "# ax[1]=ax.twinx()\n", + "ax[1].plot(range(1,26),\n", + " history.history['accuracy']+history2.history['accuracy'],\n", + " label='training accuracy',\n", + " # color=\"blue\",\n", + " marker=\"o\")\n", + "ax[1].plot(range(1,26),\n", + " history.history['val_accuracy']+history2.history['val_accuracy'],\n", + " label='validation accuracy',\n", + " # color=\"blue\",\n", + " marker=\"o\")\n", + "\n", + "ax[1].set_xlabel(\"epochs\", fontsize = 14)\n", + "\n", + "ax[1].set_ylabel(\"Accuracy\",color=\"blue\",fontsize=14)\n", + "\n", + "ax[1].legend(loc='best')\n", + "ax[1].grid(True)\n", + "\n", + "fig.show()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 286 + }, + "id": "MOnz4qCBcJVU", + "outputId": "5321280f-a0b9-4bf6-c326-7b977de4a598" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "iVBORw0KGgoAAAANSUhEUgAABCMAAAFBCAYAAABNS2U+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3hUVfrA8e/NpE1CSEgCCSRAQCCEFAhVDdXQREAERVkssKK7uKy6ughYQHFZXWEV3XX3J7BYcFVYmhQRaUGw0ROkt5AeQkmflJmc3x+ThJRJz6TA+3meeZK5c+697z3JTea+c+55NaUUQgghhBBCCCGEEA3FprEDEEIIIYQQQgghxO1FkhFCCCGEEEIIIYRoUJKMEEIIIYQQQgghRIOSZIQQQgghhBBCCCEalCQjhBBCCCGEEEII0aAkGSGEEEIIIYQQQogGZdvYAdSVp6en8vPzK36elZWFs7Nz4wV0i5J+tQ7pV+uRvrUO6VfrsUbfHj58+KpSqnW9blRYVPL9iJwn1iN9ax3Sr9YjfWsd0q/WU999W9l7kWafjPDz8+PQoUPFzyMiIhg6dGjjBXSLkn61DulX65G+tQ7pV+uxRt9qmna5XjcoKlTy/YicJ9YjfWsd0q/WI31rHdKv1lPffVvZexG5TUMIIYQQQgghhBANSpIRQgghhBBCCCGEaFCSjBBCCCGEEEIIIUSDavZzRgghhLCO/Px84uLiyMnJKV7m6urKqVOnGjGqW1dd+tbR0RFfX1/s7OzqOSpRF5bOIVE3zf1vkJyrQghxkyQjhBBCWBQXF4eLiwt+fn5omgZARkYGLi4ujRzZram2fauU4tq1a8TFxdGpUycrRCZqy9I5JOqmOf8NknNVCCFKk9s0hBBCWJSTk4OHh4dcRDVxmqbh4eEhn743QXIOiZLkXBVCiNIkGSGEEKJCchHVPMjPqemSn40oSX4fhBDiJklGlHBpyxY2Dh/OF0FBbBw+nEtbtjR2SEIIcdtKTU3lX//6V63WHTNmDKmpqZW2mT9/Pjt37qzV9svy8/Pj6tWr9bItIepLczqHhBBCNK6NR+MJe3s3077NIuzt3Ww8Gm/1fcqcEYUubdnCgQULMBUOnctOTOTAggUAdBo7tjFDE0KI21LRhdQzzzxT7jWj0YitbcX/wr755psqt79w4cI6xSdEUyfnUHlVHbcQQjSkjUfjWbz9DAmpBtq56Zk9yp8JoT6NEse89ccx5JsAiE81MG/9cQCrxiMjIwpFLl1anIgoYsrJIXLp0kaKSAghmpf6Hl02d+5cLly4QK9evZg9ezYREREMGjSI8ePH06NHDwAmTJhAnz59CAwMZNmyZcXrFo1UiI6OJiAggKeeeorAwEBGjhyJwWAAYNq0aaxdu7a4/YIFC+jduzfBwcGcPn0agJSUFEaMGEFgYCAzZsygY8eOVY6AePfddwkKCiIoKIilhf9DsrKyuO++++jZsydBQUGsXr26+Bh79OhBSEgIr7zySp36SzR/RZ9KdZq7tV4+lWpO59DMmTPp27cvgYGBLCj8MAjg4MGD3H333fTs2ZP+/fuTkZGByWTiz3/+M0FBQYSEhPCPf/yjVMwAhw4dYujQoQC8/vrrPPbYY4SFhfHYY48RHR3NoEGD6N27N7179+bHH38s3t/f/vY3goOD6dmzZ3H/9e7du/j1c+fOlXouhBC1VZQAiE81oLiZAGiIEQllLd5+pjgRUcSQb2Lx9jNW3a+khgtlJyXVaLkQQoibrDG67O233+bXX3/l2LFjAERERHDkyBF+/fXX4pnoV65cibu7OwaDgX79+jFp0iQ8PDxKbefcuXN8+eWXLF++nMmTJ7Nu3ToeffTRcvvz9PTkyJEj/Otf/2LJkiWsWLGCN954g3vuuYd58+bx7bff8p///KfSmA8fPszHH3/ML7/8glKKAQMGMGTIEC5evEi7du3YunUrAGlpaVy7do0NGzZw+vRpNE0jNja2Vv0kbg3W+FSqOZ1DixYtwt3dHZPJRHh4OFFRUXTv3p2HH36Y1atX069fP9LT0zGZTCxbtozo6GiOHTuGra0t169fr7IvTp48yf79+9Hr9WRnZ7Njxw4cHR05d+4cU6ZM4dChQ2zbto2vv/6aX375BScnJ65fv467uzuurq4cO3aMXr168fHHHzN9+vSa/iiEEKKcihIA72w/3WCjIwx5JnadTiY+1WDx9YQKltcXSUYUcvL2Jjsx0eJyIYS43R1+6y1unDmDyWhEZ2GI89XISAry8kotM+Xk8Mtrr3Gh8JPTslr5+9Nn3rwaxdG/f/9SJfE++OADNmzYAEBsbCznzp0rdyHVqVMnevXqBUCfPn2Ijo62uO2JEycWt1m/fj0A+/fvL97+6NGjadWqVaXx7d+/nwceeABnZ+fibe7bt4/Ro0fz4osvMmfOHMaOHcugQYMwGo04Ojry5JNPMnbsWIYMGVKjvhDNyxubT3AyIb3C14/GpJJnKii1zJBv4qW1UXx5IMbiOj3atWTBuMAaxdFUz6E1a9awbNkyjEYjiYmJnDx5Ek3TaNu2Lf369QOgZcuWZGRksHPnTn7/+98X327h7u5e5XGPHz8evV4PQH5+PrNmzeLYsWPodDrOnj0LwM6dO5k+fTpOTk6ltjtjxgw+/vhj3n33XVavXs2BAweq3J8QQlRGKVVJAiCH1zb+yqhAbwZ0dsdOV783M+QaTXx/9iqbIxPYeSqZ7DwTNhoUqPJt27np63XfZUkyolDP558v9akegM7RkZ7PP9+IUQkhRPNQNhFR1fLaKrrIB/OnvDt37uSnn37CycmJoUOHWiyZ5+DgUPy9TqcrHmJeUTudTofRaKzXuLt168aRI0f45ptvePXVVwkPD2f+/PkcOHCAXbt2sXbtWt5//3327t1br/sVzUfZRERVy2urKZ5Dly5dYsmSJRw8eJBWrVoxbdq0WpW/tLW1paDA3F9l1y953O+99x5eXl5ERkZSUFCAo6NjpdudNGlS8QiPPn36lEvWCCFETcTdyOblDb9W+LqjnQ3/OxzLqp8v46q3I7x7G0YGejG4W2uc7Gt3+W40FfDjhWtsjkzg2xNJZOQYaeVkx4RQH8aFtCMx1cArG38tNVJDb6dj9ij/Wu2vuiQZUahoGPHBN97AmJ2NU9u29Hz+eZm8UgghoHgEQ0ZGBi4uLuVe3zh8uOXRZW3bMvyTT2q1TxcXFzIyMip8PS0tjVatWuHk5MTp06f5+eefa7WfyoSFhbFmzRrmzJnDd999x40bNyptP2jQIKZNm8bcuXNRSrFhwwZWrVpFQkIC7u7uPProo7i5ubFixQoyMzPJzs5mzJgxhIWFlfq0Wtx6qhrBEPb2boufkvm46Vn9u7tqtc/mcg6lp6fj7OyMq6srycnJbNu2jaFDh+Lv709iYiIHDx6kX79+ZGRkYDQaGTFiBB999BHDhg0rvk3D3d0dPz8/Dh8+zL333su6desqPW5fX19sbGz49NNPMZnMb75HjBjBwoULmTp1aqnbNBwdHRk1ahQzZ86s8lYtIYSoiKlAseqnaN4pnIdhYm8fth1PxJB/M+mst9Px1sRgRgV68/25FLafSGLXqSusPxqPo50Ng7q2ZlSgN8MD2uDmZA9UPAlmQYHiYPR1Nkcl8M3xJK5n5eHiYMvIQG/G9WxLWBfPUqMubGw0Fm8/Q3yqAZ8GmkxTkhEldBo7luzERCKXLmXspk3YFg7TE0IIUTlrjC7z8PAgLCyMoKAg7r33Xu67775Sr48ePZr/+7//IyAgAH9/f+68885a76siCxYsYMqUKaxatYq77roLb29vi8mYIr1792batGn0798fMA/vDg0NZfv27cyePRsbGxvs7Oz497//TUZGBvfffz85OTkopfjrX/9a7/GL5mP2KP9Sc0ZA3T+Vai7nUM+ePQkNDaV79+60b9+esLAwAOzt7Vm9ejV//OMfMRgM6PV6NmzYwIwZMzh79iwhISHY2dnx1FNPMWvWLBYsWMCTTz7Ja6+9Vjx5pSXPPPMMkyZN4rPPPmP06NHFoyZGjx7NsWPH6Nu3L/b29owZM6b4vJw6dSobNmxg5MiR9d5HQohb37nkDOasi+JITCpDurVm0QNB+LZyYnDX1hVW0xgV6M2oQG/yTQUcuHSd704k8d3JZHacTEZnozGgkzvero58E5VIjtGc0IhPNTBnXRQbjsRxOjmD5PRcHO1sGB7gxbie7RjSrTWOdjqLMU4I9WFCqA8RERGV/g2tT5pSFm4OaUb69u2rDh06VPy8rp13acsWfpozh7FbttBSPqUq1pC/lLcT6Vfrkb6tu1OnThEQEFBqWUUjI8D89zNy6VKyk5Jw8va+JUaX5ebmotPpsLW15aeffmLmzJnFkwHWt8r6tjos/bw0TTuslOpb19hE1Uq+Hyn6+2PpZ1KZplLirT7V9zlU1/OktpYsWUJaWhpvvvlmnbdV09+LhiD/M61H+tY6mku/5hkL+HfEBT7ccx5nBx3zx/VgQi8fNE2r1faUUkTFpbH9RBLbTyRxISWrwrYjepgTEOHd2+DsUP0xCPXdt5W9F2nQkRGapo0G3gd0wAql1NtlXp8GLAaK6pn8Uym1oiFjdPLyAiA7OVmSEUIIUQOdxo5t9smHsmJiYpg8eTIFBQXY29uzfPnyxg5J3MKKPpW6ldwK59ADDzzAhQsX2L17d2OHIoRoRo7FpjJnbRRnkjMY17MdC8b1wLOFQ9UrVkLTNHq2d6NnezdeGt2dTnO3YmlogQYsf7yGn0VErYFdCxmSFgdHfSF8PoRMrlO8VWmwZISmaTrgQ2AEEAcc1DRtk1LqZJmmq5VSsxoqrrKKkxEW7n0WQghxe+natStHjx5t7DCEaLZuhXOoqBqIEEJUR3aekb9/d5aPf7hEGxdH/vNEX8IDvKyyr3ZueovzDdW4CkbUGtj8LOQb0ADSYs3PwaoJifqtE1K5/sB5pdRFpVQe8BVwfwPuv1pKjowQQgghhBBCCCGq44fzVxm19Hv+s/8SvxnQgR0vDLZaIgLM8w3py8wBUav5hnYthPwySY18g3m5FTXkbRo+QGyJ53HAAAvtJmmaNhg4C/xJKRVroY3V6BwccHB3JzspqSF3K4QQQgghhBCimSg5z4+3qyPtW+k5EH2Dzp7OrH76TgZ0tn4Z4KJb++o031B+jnkkhCVpcfUQZcWaWjWNzcCXSqlcTdN+B3wK3FO2kaZpTwNPA3h5eREREVH8WmZmZqnntVHg7EzsyZNk13E7t5L66FdRnvSr9Ujf1p2rq2u5soAmk6nSUoGi9uratzk5OfI7L4QQQjSAjUfjS1VASkzLITEth+EBbfjnb3pXWLHCGuo039C1C/C/aRW/7upbu+1WU0MmI+KB9iWe+3JzokoAlFLXSjxdAbxjaUNKqWXAMjDPXl1yts/6mP1z79q1ZMXHN4sZWhtKc5mxtrmRfrUe6du6O3XqVLlZ6xtrJvvbQV371tHRkdDQ0HqMSAghhBCWLN5+plQp5iKnEjMaNBFRJ8fXwubnwMYW7vwDHF5Z+lYNO715Eksrasg5Iw4CXTVN66Rpmj3wCLCpZANN09qWeDoeONWA8RVz8vaW2zSEEKIZatGiBQAJCQk8+OCDFtsMHTqUkiWhLVm6dCnZ2dnFz8eMGUNqamqd43v99ddZsmRJnbcjhLU09XNICCHqRdQaeC8IXnczf41aU6PVEyxMGlnZ8iYlLxs2/RHWPQlegfD7/TD6rzDuA3Btj0ID1/bm51auptFgyQillBGYBWzHnGRYo5Q6oWnaQk3Txhc2e1bTtBOapkUCzwLTGiq+kpy8vMhLT8dY4p+oEEKI5qNdu3asXbu21uuXvZD65ptvcHNzq4/QhGgWbvdzSClFQUFBY4chhLCGosoRabGAulk5ogYJCa+WjhaX17iKRUO7chpWhMORz2DgCzBtK7gV3rwQMhn+9Ct7h26EP/1q9UQENOzICJRS3yiluiml7lBKLSpcNl8ptanw+3lKqUClVE+l1DCl1OmGjK+IU1vzAA2pqCGEENX33+P/xW+pHzZv2OC31I//Hv9vnbY3d+5cPvzww+LnRaMKMjMzCQ8Pp3fv3gQHB/P111+XWzc6OpqgoCAADAYDjzzyCAEBATzwwAMYDDc/tZg5cyZ9+/YlMDCQBQsWAPDBBx+QkJDAsGHDGDZsGAB+fn5cvXoVgHfffZegoCCCgoJYunRp8f4CAgJ46qmnCAwMZOTIkaX2Y8mxY8e48847CQkJ4YEHHuDGjRvF++/RowchISE88sgjAOzdu5devXrRq1cvQkNDZd6OW1UdP6kr61Y8h7Zt28aAAQMIDQ1l+PDhJBe+V8vMzGT69OkEBwcTEhLCunXrAPj222/p3bs3PXv2JDw8vFQ/FAkKCiI6Opro6Gj8/f15/PHHCQoKIjY21uLxARw8eJC7776bnj170r9/fzIyMhg8eDDHjh0rbjNw4EAiIyOr/fMSQjSQOlaOUErh7mxXbnmtqlg0pKP/heXDIPMKPLoOhi8AXfnjaFBKqWb96NOnjyppz549qq6SDhxQ/+3RQyX+9FOdt3WrqI9+FeVJv1qP9G3dnTx5styy9PR0i20/j/pcOS1yUrxO8cNpkZP6POrzWu//yJEjavDgwcXPAwICVExMjMrPz1dpaWlKKaVSUlLUHXfcoQoKCpRSSjk7OyullLp06ZIKDAxUSin197//XU2fPl0ppVRkZKTS6XTq4MGDSimlrl27ppRSymg0qiFDhqjIyEillFIdO3ZUKSkpxfsuen7o0CEVFBSkMjMzVUZGhurRo4c6cuSIunTpktLpdOro0aNKKaUeeughtWrVqnLHtGDBArV48WKllFLBwcEqIiJCKaXUa6+9pmbOnKmUUqpt27YqJydHKaXUjRs3lFJKjR07Vu3fv18ppVRGRobKz88vt21LPy/gkGoC/6tvh0fJ9yNFf38s/UwqFLlaqb94KbWg5c3HX7zMy2vpVjyHLl++XBzr8uXL1QsvvKCUUuqll15Szz33XHG769evqytXrihfX1918eLFUrGWPA+VUiowMFBdunRJXbp0SWmapn4q8f7P0vHl5uaqTp06qQMHDiillEpLS1P5+fnqk08+KY7hzJkzqux71CI1+r1oIPI/03qkb62jTv26wLX039rih2u1Vt9wJE51nLNFPfflEXX3W7uU35wt6u63dqkNR+JqH5M15WQote5p8zF+fJ9SaQmVNq/v39nK3os0tWoaTYKztzcA2YmJjRyJEEI0Dc9/+zzHko5hMpnQ6cpPzPRz3M/kmnJLLcvOz+bJr59k+eHlFrfZy7sXS0cvrXCfoaGhXLlyhYSEBFJSUmjVqhXt27cnPz+fl19+me+//x4bGxvi4+NJTk7Gu/Bvd1nff/89zz77LAAhISGEhIQUv7ZmzRqWLVuG0WgkMTGRkydPlnq9rP379/PAAw/g7OwMwMSJE9m3bx/jx4+nU6dO9OrVC4A+ffoQHR1d4XbS0tJITU1lyJAhADzxxBNMmjSpOMapU6cyYcIEJkyYAEBYWBgvvPACU6dOZeLEifj6Wnd261uBpmmjgfcBHbBCKfV2mdc7AiuB1sB14FGlVFzha+8A92EeQboDeK7wDVXtbZsLSccrfj3uIJQ5h8g3wNez4PCnltfxDoZ737b8GrfmOZSQkMCMGTNITEwkLy+PTp06AbBz506++uqr4natWrVi8+bNDB48uLiNu7t7hXEV6dixI3feeWelx6dpGm3btqVfv34AtGzZEoCHHnqIN998k8WLF7Ny5UqmTZtW5f6EEI2gZVtITyi/vBqVI1Iycnl98wlCO7jx98m90NloVgiwHiUdh/9Nh2vnYeg8GDwbbJrOBJsNeptGc6Fv0waQ2zSEEKK6yiYiqlpeXQ899BBr165l9erVPPzwwwD897//JSUlhcOHD3Ps2DG8vLzIycmp8bYvXbrEkiVL2LVrF1FRUdx333212k4RBweH4u91Oh1Go7FW29m6dSt/+MMfOHLkCP369cNoNDJ37lxWrFiBwWAgLCyM06cb5S7GZkPTNB3wIXAv0AOYomlajzLNlgCfKaVCgIXAW4Xr3g2EASFAENAPGGL1oCs6V+QcKmX27NnMmjWL48eP89FHH9Vqf7a2tqXmgyi5jaIkCdT8+JycnBgxYgRff/01a9asYerUqTWOTQjRADwruJWi491Vrvr6phNk55pY/GBI005EKAWHVsLycMjNgCc2wdC5TSoRAQ1b2rPZ0Dk44ODuLhU1hBCiUNEIhorKT/ot9eNy2uVyyzu6diRiWkSt9/vwww/z1FNPcfXqVfbu3QuYRxW0adMGOzs79uzZw+XL5fdb0uDBg/niiy+45557+PXXX4mKigIgPT0dZ2dnXF1dSU5OZtu2bcXlYF1cXMjIyMDT07PUtgYNGsS0adOYO3cuSik2bNjAqlWranxcrq6utGrVin379jFo0CBWrVpFWFgYBQUFxMbGMmzYMAYOHMhXX31FZmYm165dIzg4mODgYA4ePMjp06fp3r17jfd7G+kPnFdKXQTQNO0r4H7gZIk2PYAXCr/fA2ws/F4BjoA9oAF2QN0/nahkBANgniMiLbb8ctf2MH1rrXd7q51D6enp+Pj4APDppzdHjIwYMYIPP/yweA6KGzducOedd/LMM89w6dIlOnXqxPXr13F3d8fPz48tW7YAcOTIES5dulThviwdn7+/P4mJiRw8eJB+/fqRkZGBXq/H1taWGTNmMG7cOAYNGkSrVq2qfVxCiAZy/SJE74NOQ8zfp8VBSx9w9oSo1aBvBaP+avGifdvxRLYeT2T2KH+6tGnCZc5z0swlO09sgDvugQeWQYvWjR2VRZKMqICTt7eMjBBCiGpaFL6Ipzc/TXb+zdnzneycWBS+qE7bDQwMJCMjAx8fH9oWTi48depUxo0bR3BwMH379q3yonzmzJlMnz6dgIAAAgIC6NOnDwA9e/YkNDSU7t270759e8LCworXefrppxk9ejTt2rVjz549xct79+7NtGnT6N+/PwAzZswgNDS00lsyKvLpp5/y+9//nuzsbDp37swHH3yAyWTi0UcfJS0tDaUUzz77LG5ubrz22mvs2bMHGxsbAgMDuffee2u8v9uMD1Dyyj4OGFCmTSQwEfOtHA8ALpqmeSilftI0bQ+QiDkZ8U+llMVS45qmPQ08DeDl5UVERARgnkwxIiICV1fXak82ahv2Eo7fvYRmvDmpmrLVkxP2EsY6TFjaoUMH0tLS8Pb2pkWLFmRkZHD//fczefJkAgMDCQ0NpVu3bmRmZhbHmpGRQWZmJgUFBWRkZPDoo48yc+ZM/P398ff3p1evXmRlZdG7d2+CgoLo1q0bvr6+DBgwgJycHDIyMnj88ccZOXIkbdu2ZevWrSilyMzMpGvXrkyZMoW+ffsC8Pjjj9OlSxcuX75cvD+A3NxccnNzy/XfnDlzePDBB3Fzc2Pw4MGYTCYyMjJ47rnnePHFF+nRowc6nY65c+cyfvx4li5dyoQJEygoKKB169Z8/fXXjBw5kpUrVxIQEEDfvn3p0qULmZmZAKVi6Ny5s8Xjy83NZeXKlTzzzDPk5OTg6OjIpk2baNGiBd26daNFixY8/PDDFf7sc3Jyin9Xmoqi31lR/6RvraO2/Rpwcgme6PjFexp5HUvcuqVM3KH7hPa//B9Xzx/lZI8XKNDdrJiRmad4eb+Bji1t8FexRETE1cNR1I82yXvpfHEVDrlXybNzRVMF2BkzudTpMWJ8J8KhEzXaXkP+zmp1vQWysfXt21eVrHUdERFRnJWvi72zZpEVH8+YDRvqvK1bQX31qyhN+tV6pG/r7tSpUwQEBJRaVtHICDBX03hl1yvEpMXQwbUDi8IXMTVYhilXV2V9Wx2Wfl6aph1WSvWta2zNjaZpDwKjlVIzCp8/BgxQSs0q0aYd8E+gE/A9MAnzbRmemBMUDxc23QG8pJTaV9k+S74fKfr7Y+lnUqmoNebZ3NPizPcuh89vkNJqzUldzxNrS0hIYOjQoZw+fRobG8t3Q9f496IByP9M65G+tY5a9WvCUVg21Dxvwj2vWm7zyzL4dg607QlTVoOLFwAvrD7GpsgENs0aSI92LesUe70qKlNaqjqIBkNegmEv12qT9f07W9l7ERkZUQEnb29SDh9u7DCEEKLZmBo8VZIPoqmIB9qXeO5buKyYUioB88gINE1rAUxSSqVqmvYU8LNSKrPwtW3AXUClyYh6ETJZkg/N2GeffcYrr7zCu+++W2EiQgjRiHa+Dnp3uPvZitsMeBrc2sPa38KK4TD1f+y+3or1R+N5Nrxr00pEAHz3WvkypSg49kWtkxENSf5SVsDJy4u89HSM2dlVNxZCCCFEU3IQ6KppWidN0+yBR4BNJRtomuapaVrR+6B5mCtrAMQAQzRNs9U0zQ7z5JUWb9MQoqTHH3+c2NhYHnroocYORQhR1oXdcDHCPGLAsYqEgv+9MP0bMOWi/jOCdWu/pJtXC2YN69IgoVYp+zocWA7L74HMCuY4TGs6t5FURpIRFXAqvDdZ5o0QQgghmhellBGYBWzHnEhYo5Q6oWnaQk3Txhc2Gwqc0TTtLOAFFE1wsha4ABzHPK9EpFJqc0PGL4QQoh4VFJhHRbh1gL6/rd467UJhxk6S8eC9/IUs73kee9tGvHQ25cOZbbD6Mfi7P3zzZ8jPAUc3y+2rUaa0KZDbNCrg5GW+Pyg7OZmWhfWphRDidqOUQtOacOkqAZh/TqI0pdQ3wDdlls0v8f1azImHsuuZgN/VYxxyDolicq4K0QhOrIfESHNVCVuHqtsX+uGqEzPTXmZTm4/w+/4FsEmBIXOgIf+mJx0333Jx/H+QlQJOntD3Seg1BbxDzMvLzhlhpzfPOdQMSDKiAs7e3gBkJyY2ciRCCNE4HB0duXbtGh4eHnIx1YQppbh27RqOjo5VNxYNSs4hUZKcq0I0AmMe7H4TvIIhuPq3UGXlGpm7PgoPzzZ4P7MFvn0RIt6CG9Ew7gOwta+f+CxNXNx5qDnJcOxLSD4ONnbgPxp6/ga6jgCd3c31i+YZaqaTH0syogL6EiMjhBDidqcC5coAACAASURBVOTr60tcXBwpKSnFy4rK2In6V5e+dXR0xNe3eQzJvJ1YOodE3TT3v0FyrgrRwA5/Yk4gTF0HNZhYdvH2M8TdMLDmd3fh6KiH+z+EVn6wZ5H5ov/hVaBvVbfYylbCSIuFDb8DpQAF7XrDmCUQNAmc3CveTjOe/FiSERXQ2dvj6OFBdlIFk4IIIcQtzs7Ojk5lblOLiIggNDS0kSK6tUnf3nosnUOibuQ8EUJUW24G7P0b+A2CLuHVXu1g9HU+/SmaJ+7yo59fYRJAKyyX6dYRvv4D/GcU9JkGP/+r+iMSlDJPPpkWA6kxhfM+lKmEoQrAwQWe3Altutf4kJsbSUZUQu/lJSMjhBBCCCGEEKK5+fGfkH0Vhr9R7XkecvJNzFkbhY+bntmj/Ms36PkwuPrA5w/C9nk3l6fFmkc55GRAu143Ew7Fj1jz1/ysqoPIzbwtEhEgyYhKOXl5kRUfX3VDIYQQQgghhBBNQ+YV+Omf0ON+8O1T7dXe23mWi1ez+PzJATg7VHCp7DcQHF0hs8yohnwDfPNC6WWOruYqHh53mOeCcOtQ+GgPX06BdAvXms2kEkZ9kGREJZy8vUk5fLixwxBCCCGEEEIIUV3fLzYnB+6pflWJyNhUln9/kUf6tWdgV8/KG2dWMnr+kS9vJhwcXStuN/z1Zl0Joz5IMqISTl5e5KWnY8zOxtbJqbHDEUIIIYQQQghRmesX4dBK6PMEeHap1iq5RhOz10bSxsWRl+8LqHoFV1/zrRnllreH7mOqF2czr4RRHyQZUQmntm0Bc0WNljIBlRBCCCGEEEI0bbv/Ajp7GDKn2qt8uOcCZ5Mz+c8TfWnpaFf1CuHz62dUQzOuhFEfql/f5DbkJOU9hRBCCCGEEKJ5SDgKv66DO58BF+9qrXIyIZ1/7TnPA6E+hAd4VW8/IZNh3AfmkRBo5q/jPritEwu1ISMjKuHsbf4Fzk5MbORIhBBCCCGEEEJUaufroHeHsGer1dxoKuCldZG4Odkxf2yPmu3rNh/VUB8kGVEJvYyMEEIIIYQQQoim78JuuBgBo96qfOJIYOPReBZvP0N8qvk2i+l3d6SVs30DBClKkts0KqGzt8fRw4PspKTGDkUIIYQQQgghhCUFBeZREa4doN+TlTbdeDSeeeuPFyciAL46GMfGoxbKbAqrkmREFfReXjIyQgghhBBCCCGaqhPrITES7nkVbB0qbbp4+xkM+aZSywz5JhZvP2PNCIUFkoyogrO3t4yMEEIIIYQQQty+otbAe0Hwupv5a9Saxo7oJmMe7H4TvIIh+KEqmyeUGBFRneXCeiQZUQW9l5ckI4QQQgghhBC3p6g15jKWabGAMn/d/GzTSUgc/gRuRMPwBWBT9eVtS73laRPbuenrNy5RJZnAsgpOXl7kpadjzM7G1smpscMRQgghhBBCiIazayHklxk1kG+Ab/5sviXCtT24dQQnd9C0ircTtca8rbQ4cPWF8Pl1r0aRmwF7/wZ+g6DL8Cqbf3ciiTSDERsNCtTN5Xo7HbNH+dctFlFjkoyoglPbtoC5okbLTp0aORohhBBCCCGEaEBpcZaX56TBmsdvPrdzBrf24NbB/HAt+r4jJByFHa/eTGoUja6AuiUkfvoQsq/C8DcqT4QAkbGpPPvVUXr6ujJ1QAfe33WehFQD7dz0zB7lz4RQn9rHIWpFkhFVcCoq75mUJMkIIYQQQgghxO0j+aT5Il+p8q+19IHfrIbUmPKP2AOQk1r5tvMNsOuN2icjMq/Aj/+AHveDb59Km8Zez+bJTw/h2cKBFU/0o7WLA5P7dajdfkW9kWREFZy9vQFk3gghhBBCCCHE7SP5BHw6DuxdwJQLxpybr9npYfjr4B1sfliSk24eAZEaA18+YrlNWhysfhQ6DgS/MGgTWK15HwD4frE5oXHP/EqbpRnymf7JQfKMJr56egCtXSqvtiEajiQjqqAvHBmRJckIIYQQQgghRDUc3PQR7Y8sZrBKISmiNbG9Z9Nv/O8aO6zqSzoOn44HW0d4cgfEH675fA+OLcExELwCzbdspMWWb2PnZC7JeWqz+bm+FXS4G/wKkxNeQWCju9m+cN6JIUXb8hsMnl0qDCHPWMDvVx3m8rUsPv1tf7q0calhRwhrkmREFXT29jh6eGBITm7sUIQQQgghhBBN3MFNHxF0+FX0Wh5o4E0Krodf5SA0j4REYhR8Nt48B8S0zeDeGTzuqNvcDuHzzXNElJwI004P4943bzc1BqJ/gMv7IXo/nNlqbuPoWpicCDOvu/9dyDdQPDtE3AFzgsJCbEop5q0/zk8Xr/H3h3py9x2etY9fWIUkI6pB7+VFtiQjhBBCCCGEaLrqo1pDXbahFGReofORv5oTESXotTzaH1kMTT0ZkXAMPrsfHFzgic3gbp4zb+PReBZvP1P7CR+L+rCivnXrAL06QK8p5udp8XD5B4jeZ05SnN1mebvGHPM2LfyMPth1nnVH4nh+eFcm9fGtfqyiwUgyohqcvb3JjKtgFlkhhBBCCCFE44paU/qT99pUa6hqGwUFkJkEqYXzIKRevjknQmqMebkpF48KNt9GXa314dVWjZIICUcLExGu5hERrfyKtzFv/XEM+SYA4lMNzFt/HKBGCYmNpjAW535AQo6Bdo56Zpv8mVBRY1cfc58X/ezSE+Hd7pbbWqj2se5wHO/tPMvE3j48F9612jGKhiXJiGrQe3lx5dChxg5DCCGEEEIIYcmuhaVvAQDz821zgMpLPhbbNsfyNjbNgoi3zBe9ptIjHnDyBLf2FLQJ5JRLGF9H63harcVTSy+3+XzNFruL+7DpPKj6x1UHVSUR8owFZOYaycwxkh97iI5bf0O+vSv7+v+HK2c0MnIukJmbzyc/RBdvo4gh38Trm0/g4miLu7M9Hs4OuLewx9leh2ahxGadExot21Y874Rr6VEPP164ytz1UdzV2YO3J4ZYjEc0DZKMqAZnb2/y0tMxZmdj6+TU2OEIIYQQQgghSrLw6TgAhuuwfkbdtm3MhbY9IWCc+XYCt47mC2O39mDvzC8Xr7Fg0wlOJ2UQ1sWDYy6dCTu5sNStGnlKRw62OHw2lpx2d+I4/GXoNNhcNtNKFm8/YzGJ8MKaY7y0Loo8YwEAvbTzfGb/FgmqBVMyZhO/OQVIAUBno2EqsFDWE0jNzufJT0t/YGtva4OHs705QdHCofj7/x2KtRjL4u1nqj+6oqJ5J8JvVtM4l5zB71Ydxs/Dmf97rA/2ttWszCEahSQjqkFfVN4zOZmWnTo1cjRCCCGEEEKIkgyOrdHnXLGwvA36GRXMN1DWp/dBhoUKeq7t4aFPyi1OSsvhr+uOsikyAR83Pf+a2pt7g7zRtDs5uMme9kcW00Zd5YrmSWyfPxPTJpyz2/7Jk/Ff4/jZeFSHu9CGzIHOQ+s9KWE0FRCfarD4WoGC34Z1wsXRFj/DSUYeeYd8B0+ujPyCf3t0oIWDLS0cbXFxsMPRzoaBf9tjcVteLR346LG+XM/K5VpmHtez8riWlVf4fS7Xs/K4mJLJ9aw8svNMFiKBhApitKjEvBMqLQ6tzLwTKRm5TP/kIA62OlZO64er3q762xaNQpIR1eBUWN4zOylJkhFCCCGEEEI0JakxGHJycFSlr+mzlT3vGH/D65WUfixlxJtVfvIOkGs0sXJ/NP/YfQ5jgeLZe7owc2gX9PY3S1D2G/87GP87IiIiGDp0KN5APyC221u8sGYinWPX8Xz8VtxXTYD2A2DIS3BHeJ2TEqYCxebIBN7fda7CNj5ueube2x1ifoHPn4GWbbB7Ygt9XS2PUJg9yr/ULRYAejsd8+4NoFd7t2rFdfdbu0hIyym33N7Whj1nrjC0W+vq3U5ROI/E3sJ+LWLIMzHj04Nczcxl9dN30d5dRrM3BzJupRqci0ZGJFnIlAohhBBCCCEax43L8Ml96JSRvxsfJK7AkwKlEVfgydz8GXyS2Z9jsakV3mpQSshkGPeBeSQEmvnruA9KTYC558wVRi/dx9++Pc3dd3iy809DeGGkf6lERGXauzvx2dOD6TD6eQblvsdftacwXI2BzyfBiuFwboe5KkcNFRQmIUa+t5fnVx/DwdaGJ8P80NuVvtzT2+mYPcofYn6GzydCizYwbat5wsgKTAj14a2Jwfi46dEwJzPemhhco8krXxrdHb1d6T6ytdFwsNWY/vFBRi39njWHYsk1Wh5BURlTgeK5r44SFZ/GB4+E0rOaCRLR+GRkRDXoC0dGZEkyQgghhBBC3CoKy1gOSYuDo7UshdmYbkSjPrmP/OwMHs9/mciCzvzTNLFcswkf/oCr3o6BXTwZ3M2TQV1b085Nb3mbJSs4lHD5WhZvbjnJzlNX6OTpzCfT+zHUv02twrax0ZgxqDNDurXmT2ta8Un8QBb5RTEpYzU2/30Q2vWGIXMgN73KMqMFBYrtJ5JYuvMcZ5Iz6NqmBf+a2pvRgd7Y2GgE+7qVr6bhfhlWPQgt25nLd7ZsW2XME0J9albK08L6QLlYxgS3ZXNkAsv3XeSltVEs2X6GaWF+TO3fEVen6t1msWjrKb47mcyCcT0YGehd6xhFw5NkRDXo7O1x9PDAkJzc2KEIIYQQoho0TRsNvA/ogBVKqbfLvN4RWAm0Bq4Djyql4gpf6wCsANoDChijlIpuuOiFaAAlylhqULtSmI3p+iXy/jOG3OwMHsmZR5prd+wz84onZQTzKIBX7gugpd6OfWdT+P5cCluPJwLQpU0LBnX1ZHDX1gzo7I6TvfmyqGwpzOfCuxJ7I5uPvr+IrY3GnNHd+e1APxxsqzcSojJdvVzY8EwY/9h9nrl77PmgRX9WDrhA1zP/B18+jLkKSOEoiTI/H6UUO04m897Oc5xKTKdza2c+mBLKfcFt0dncvN2hXBIhej98Ptk8EuKJzeDScBfvFSU0JvXxZWJvH/adu8ryfRd559sz/HP3eR7u157fhnWq9JaLj3+4xMofLjE9zI/pYXI7fXMjyYhq0nt5ycgIIYQQohnQNE0HfAiMAOKAg5qmbVJKnSzRbAnwmVLqU03T7gHeAh4rfO0zYJFSaoemaS2AAoS41VRUCnPXwiafjLgWcwrdZ+MoyDcwy3YBT0y8l0l9fNkcmVB+FEDhxe/4nu1QSnHuSibfn01h37mrfHkgho9/iMZeZ0Nfv1a4O9uz42QyuYUJjfhUA3PWRaGA+3u1Y969AXi7OtbrsdjpbHhhRDfCu7fhT2uOMWKvH7+963NeO30/muFG6cb5BtTWFzkTm8LnpwvYd9UJ+1btee/hnozv6VMqCVGscPQLaXHg7AmGVHDvXJiI8KrXY6kLTdMY3K01g7u15mRCOiv2XWTVT5f59MdoxgS35enBnQnxdStOFsWnGnDft4PrWXmM6OHFq/f1aOxDELUgyYhqcvb2JjOugpJBQgghhGhK+gPnlVIXATRN+wq4HyiZjOgBvFD4/R5gY2HbHoCtUmoHgFIqs6GCFqJBVVQKMy0WfvkIgh4EZ4+GjakKuUYT63d8z7Cff4tGPuuD/81HY+/FxdE8nL+qWwk0TaOblwvdvFyYMagzOfkmDkXfYN+5FPaeTeHHC9fKraMAzxb2vP9IqLUOC4Ce7d3Y+sdBvLP9NCt/iOZVx1QsTueYm073gy/zFwAHUNka2m5vONKhsOxoh8Kyox0g+STsWQTGwqRTVgqgQf+nm1Qioqwe7Vry7sO9mD3an09+iOaLX2LYEpXIHZ7OxN4wkGcyJ4uuZ+WhaTAioI3lRIxo8hp0AktN00ZrmnZG07TzmqbNraTdJE3TlKZpfRsyvsrovbxkAkshhBCiefABYks8jytcVlIkUHRz+QOAi6ZpHkA3IFXTtPWaph3VNG1x4UgLIW4trr6Wl9vYwbaX4O/+8NVUOLUFjHkNG1sZSim+O5HEk3//kmE/T8dJV0D2lI3MeHB8cSKiNhztdAzs6sm8MQF8+/xgyxf/wLXMhjl+vb2OBeMC+WLGAJKwnAhKwpNNQ7/F+NhmmPBvtKFz4Y57QGcPsQdg/1LY8rx5csodr95MRBRT8MNS6x9MPWjrqmfemAB+nHcPr94XQPS1rOJERBGl4P1d5xspQlFXDTYyoppDJtE0zQV4DviloWKrDmdvb/LS0zFmZ2PrJKVihBBCiGbuz8A/NU2bBnwPxAMmzO+NBgGhQAywGpgG/KfsBjRNexp4GsDLy4uIiAgAMjMzi78X9Uv6tv60afcQ/un/QKfyi5eZbBw44/8Hspw74p20G68Le7E/vYU8u5ZcaTOYJO97yGzRuc7lJ2siPqOAL07nknM9lq8cFuFsq/i191/ITsrkfFJEve7L3VHjWk75Shbujlqtf+9q+zv7Tx7mVbUcJ+1mIiRb2fMP7WFGksv+WIB25ofbnVBYQEIrMGGfdx3HnCv0OvayxQSLSotjbzM7j7oApgqKjMSnGuTvQj1qyL+zDXmbRnWGTAK8CfwNmN2AsVVJX1TeMzmZlp1kchQhhBCiCYvHPPlkEd/CZcWUUgkUjowonBdiklIqVdO0OOBYifcrG4E7sZCMUEotA5YB9O3bVxXVvI+IiKDoe1G/pG/r01BYfRFObUQBmmt7dOHz6VE8X8Q0MBnhwi7sj32B75lv8I3fAm16QM8p5nkliiY/LDkvQQVVH6pSduLIPwy7gzNJGXz+SwxB9olscXkbRzt7tCc2079N93rsh5tec41n3vrjGPJvlpfU2+l47f5ghtaykkRtf2enf5tFpo3iJds1tNOukaA8eMc4mc0FYfy1utu79G/zbTdlaK6+zfI88vl5N/GpZUd6mEuNNsfjaaoa8u9sQyYjLA2ZHFCygaZpvYH2SqmtmqY1qWSEc1EyIilJkhFCCCFE03YQ6KppWifMSYhHgN+UbKBpmidwXSlVAMzDXFmjaF03TdNaK6VSgHuAQw0WuRANKTMJ2vZir/8bli8+dLbQbZT5YbgBv66HyC9hx2uwcwHcEQ7ufnDk85u3A9SiKsfGo6WTAPGpBl7e8CsAL/Yq4A8xb2FjY2uedLG1fx0PumIVlZ+sS0nL2mrnpmdT6kA25Q0stdynopKkloTPL66YUsxOb17eDM0e5W8xWTR7lPV+J4R1NZkJLDVNswHexTwUsqq2FodFgvWGlRivXgXgSEQETrm59b79pk6GRVqH9Kv1SN9ah/Sr9Ujf1h+llFHTtFnAdsylPVcqpU5omrYQOKSU2gQMBd7SNE1hvk3jD4XrmjRN+zOwS9M0DTgMLG+M4xDCqnLSIO4gDHy+eu31raDfk+bH1XPmpETkaji/o3zbalblyDWauJ6Vx6JvTpW6wCxyZ4tk/hizCGxsYdoW8OxavVjroKpJMBtKvVx4F/V/HUetNBUlk0XxqQZ8GjFZJOpHQyYjqhoy6QIEARHm//14A5s0TRuvlCr1iURFwyLBesNKTHl5rF6wgPZubgTfhsOAZFikdUi/Wo/0rXVIv1qP9G39Ukp9A3xTZtn8Et+vBdZWsO4OIMSqAQrR2C7uBWWCLsPhUg0naPTsar6oHfYqamGrCuYliOWnLZ9wTh9MQp4T1zPzuJ6Vx7Us89frWXlk5hor3IW/FsOH+YtA3wKe2AKeXWoWYzNXb6M0QiY32+SDJUXJIvmfeWtoyGREpUMmlVJpgGfRc03TIoA/l01ENBadvT2OHh4YkpMbOxQhhBBCCCHq5sIu8m2dCf8ik5i0fHx+3l3pxa5SiquZecTdyCY+1UDcDQNxN7KZqTzx0a5aaA93H3qOu4HTqgPHdYGc1ffCwTWUjh3a4O5sj4ezPe7ODiz57gzXs24mRAK0y/zXfhFGzQGmbQWPO6zVC01aUxmlIYS1NFgyoppDJps0vZcXWVLeUwghhBBCNGdKkX1yOz/m9yAm01xNIz7VwNx1UVy6mkXn1s7E3TCUSjrE3zCQayxdVtFVb0d6/mTetltRrurDq/m/ZfaU0bRKOYh/3I90j90LWdsgC8gPAJcwaBMGfgNxsu/B/g3/4nm+op12FYVGOk5Ehn/OiNs0ESHE7aBB54yoashkmeVDGyKmmnD29iYzLq6xwxBCCCGEEKL2rp7DyZDIHuO9pRbnGAt4f9e54ufuzvb4ttLj7+VCePc2+LZywsdNj6+7Hh83PS6OdoS9bcvcdMpVfTjccgRtg4cBw8wbM+VDwjGI3geXf4DIr+DgCgAmtPBmvC4FG1U0P4KipY2REa7yvluIW1mTmcCyOdB7eZF8qEncNSKEEEIIIUSN5eSbOLt3HSHA3gLLU6Ps+NNg2rnpcXao+lLBPNFiXqmqD3o7HW+VnWhRZwft+5kfg14wlw1NjITL+2HPX0skIgqbF+RWaxJMIUTzJcmIGnD29iY/PZ38rCzsnJ0bOxwhhBBCCCGq5WxyBl8eiGH9kXiWGrdx0aYtcapNuXY+bnq6erlUe7u1nmhRZwu+fcyPHQsst0mTkRFC3MokGVEDem9vALKTk3Ht3LmRoxFCCCGEEKJiOfkmtkYl8uWBGA5dvoGdTuO+AHcGXzpDdIdJ6M/r6lY6slCdJ1p09YW0WMvLhRC3LElG1IBzYTLCIMkIIYQQQojbzsaj8XUvtdgATiel89WBWNYfiSM9x0hnT2deGRPAxN4+eCT/AOdzuOOuCbwVEsji7WeITzXg05jHEz4fNj8L+Yaby+z05uVCiFuWJCNqwMnLC4BsqaghhBBCCHFb2Xg0nnnrjxePJIhPNTBv/XGARrmAL5sYeS68K5oGXx6I4UhMKvY6G+4N9mZK/w4M6OSOpmnmFX/YBTp78Atjgr0zE0J9iIiIYOjQoQ1+DMWK5oXYtdB8a4arrzkRIfNFCHFLk2REDegLkxFS3lMIIYQQ4vayePuZUrc0ABjyTSzefqbBkxGWEiMvrYsC4I7Wzrx6XwATe/vi7mxffuXzu6DDXWDfxOY/C5ksyQchbjOSjKgBnb09jh4eGJKTGzsUIYQQQojbR9Qa2LWQIWlxcLRxPjVPSDUw3mZ/YQnLqyQoT94xTmZz6sCqV65nf/v2dLnECIBnC3t2vjDk5iiIstLiIeUU9Jpi5QiFEKJqkoyoIb2Xl4yMEEIIIYRoKFFriucT0MA80eHmZ82vNWBC4nHnA8wxrsBJywPAV7vK23YrIB8eWebOtLv9GB7gha3OxmoxnEhI48sDMSSm5Vh8/VpmXsWJCIALu81fuwy3QnRCCFEzkoyoIWdvbzLjpMyQEEIIIUSD2LWw9MSGYH7+3SvQeRg4e0JlF+BFCkdX1HhOgtwM8q5F86eCj4sTEUWctDwW6b/gd1e78/zniXi4uTH1zg480q+D5VskahFLVq6RzZEJfHkghsi4NBxsbdDb6SyOjGjnpq/8WC7sApe20KZHlYcthBDWJsmIGtJ7eZF86FBjhyGEEEIIcXtIq+BDoMwrsKQL2OrBrT24dbj5cG0Pbh3N37doA8f/V7paQ8nRFd1GQWoMpMYWfo2BtJib3xtuYA9UlFpwMaXyhelZcIS0PFcu7fbgl91t0Lf2o5t/IO38/G/G5NCi1EiPcrGUSEj8Gp/GFwdi+PpoPFl5Jrp5teD1cT14INSXPWeulJozAqpRlrPABBf2QPf7qpe8EUIIK5NkRA05e3uTn55OflYWds5NbOIfIYQQQohbTQsvyLRwi6yTJwx5qTBpcNmcTIg/AobrpdvZOkKB0fwoKd8A658GVOnldk6FyYwOmNr1YeMlHXuSHFnc4gv0edfKx+HcBkb9FVIv45oaQ9crl/BJuUTLq4dxuPYV/Fiird4dcjOgIL98LLsWkuk/kU3HzKMgjsen4Whnw9iQdkzp34HeHdyKb8EomjCzRmVG449ATip0Ca+4jRBCNCBJRtSQ3tsbgOzkZFw7d27kaIQQQgghbmH5OWBj4e2qnR5Gv2X51obcDHNiIi32ZqLix39UsAMFI94sMbKiIzh5gKaRbyrg2S+Psi0hiTfGB6JvEVJ6RENRHKMWQchDxYucCx9pWbms/ekY+w4exS4jlu76Gwx2NRCUtB5L4xJUWhz9F+0kO89Ed28XFt4fyP29fHDV21mMfEKoT82qeFzYBWjmW1uEEKIJkGREDTkXJiMMkowQQgghhLCu716F9Di4649wciMqLQ6tqvkeHFzAq4f5UeTERnNyoizX9hD2bLnFRlMBf1p9jG2/JvHa2B48cbcf4Gd+sZpzPbg6OzB1+ACm3NOfiLNX+OTHyyw+m8IPDt/jo10t1z5BeTA2pC1T+negV3u3yieirI3zu8CnNzi51+92hRCiliQZUUNOXl4AUlFDCCGEEMKaTm2Gg8vhrlkw6i8w6i/sjYhg6NChNd9W+HzLoxrC55draipQvPi/SLZEJfLymO48ObDTzRdDJte4goeNjcY93b24p7sXF1IyWfqPR3iDZaUmw8xW9iyzfZR3HuxZ40OrFsMNiD8Eg/5sne0LIUQtWK/20C1KX5iMyJZkhBBCCCGEdaTGwtezoG0vCF9Q9+2FTIZxH5hHQqCZv477oFxiwVSgmL02kq+PJfDSaH+eHnxH3fddwh2tW7A2727m5s8grsATpcCkNF7Nn85nWf3rdV+lXIwAVSAlPYUQTYqMjKghnb09jh4eGJKTGzsUIYQQQohbj8kI62aYqz88uBJsKymRWRNVjGooKFDMWx/F+iPxvDiiG88M7VI/+y2jnZueTakD2ZQ3kIE2x/nc/i0UNlWX5ayL87vAwRV8+lhvH0IIUUMyMqIWnLy95TYNIYQQQghr2Ps2xP4M45aCR/2OTKhIQYHilY3HWXMojufCu/LH8K5W29fsUf7o7XQA/FAQyIWCtkyz21F5Wc66UMqcjOg8BHTyOaQQoumQZEQtOHl5ycgIIYQQQoj6dnEvfL8EQh+F4AeLF288Gk/Y27uZ9m0WYW/vZuPR+HrbpVKK+Zt+5csDsfxh2B08P9x6iQgwcklxuwAAIABJREFUV8F4a2IwPm56wIav7cfSUzvPhNZW+qAr5TRkJMgtGkKIJkfSo7Wg9/Ii+dChxg5DCCGEEOLWkZkC658Cz65w7zvFizcejWfe+uMY8k0AxKcamLf+OEDNSltaoJTijc0n+fznGH43pDN/Hulf/1UsLChVljN3MPz9KzjwEfguq/+dnd9l/tolvP63LYQQdSAjI2rB2dub/PR08rOyGjsUIYQQQojmr6AANs4EQ6p5ngh75+KXFm8/U5yIKGLIN7F4+5k67VIpxV+2nuKTH6OZMbATc0d3b5BERDkOLtDrN/Dresi8Uv/bv7ALPP3NpUiFEKIJkWRELei9vQHIlls1hBBCCCHq7ucP4fwOGLUIvINLvZSQarC4SnyqgRfXRPL5z5c5mZD+/+zdeVjVZf7/8ed9DjsqKsqiiKJpqbmj2eKEqVkTbU4/W6zGajKbasZpqrEsTcupbLNdmbIms/qabWaLoxZlq2impuUCiqK4KwrIeu7fHwcRBBQQzmF5PbrOdc65P/fnPu9zd9UFb+77fVNQ6Kr0x1lrefzz33nt282MPqcDEy7p6p1ExFEDbgVXPqx4o2bHzcuGLd9pVYSI1EnaplENwUXJiCO7dhHSsaOXoxERERGpx7avgMUPwxnx0P8vpS4dzsnHz8dBbkHZREOAj4OvN+zm/Z/TAAjyc9IzKoQ+0S3oG92CPtHNadXEv7j/Ryu38+TC9ew4eIRgfx8ycwu4YWB7Jl3azbuJCHBvTel0ASyfBef9A5y+NTNu6vdQmKtkhIjUSUpGVENQeDiATtQQERERORU5GTDvZmgaCZe/CCWSAnsO53LTG8vIK3Dh6zTkF9ria4G+Th4b0YPLe7ch7cARft56gJVbD7Jy6wH+800KBS5333YtA+kb3QKngU/X7CxOamTmFuB0GPq2a+79RMRRA26Dd66G3z6BM0fUzJjJS8AnANqfWzPjiYjUICUjqiGwKBmRrWSEiIiISPVYC5+Mg4Pb4KbPILBF8aWt+7K5YdZP7D6Uy6yb+pORnc+TC9ez/eAR2jYP5N7hpxcXgGzXMoh2LYO4vLf7fU5+Ib9uz3AnJ7Yd4KeU/ew8lFPm4wtdlqcWbeDKfnWklkLnYdCiAyxLqLlkxKYl0P4c8A2smfFERGqQkhHV4PTzIyA0VMkIERERkepaORvWfgAXPATRA4ubf92ewejXkyhwuZhz61n0jXYnKa7o05bExETi4uJOOGyAr5PYDi2J7dCyuC1m/KfYcvpWVI/CKxxO6H8r/G8CpK+GyJ6nNt7BbbB3PfT7c83EJyJSw1TAspqCIiJUwFJERESkOnb/Dp/dBzHnu2skFPk+eS/XJPyIn9Mwb+zZxYmIU9WmefkrAypq95o+14NvkPuYz1OVXHSkZyfVixCRuknJiGoKCg/niJIRIiIidZIx5iJjzHpjzCZjzPhyrrc3xiwxxqw2xiQaY6KOu97MGJNmjHnRc1E3EvlHYN5N7uM7RyS4VwQAn61JZ/SsJNo0D+D9v57DaWFNa+wj7x1+OoG+zlJtgb5O7h1+eo19Ro0IbA49r4Y18yB7/6mNtWkJNGsLrevYdxQRKaJkRDUFRUSogKWIiEgdZIxxAi8BFwPdgGuNMd2O6/YU8Ka1ticwBXjsuOuPAN/UdqyN0sIJsHsdXDkTmrpPKJv9wxbuePtnekaF8N5t5xAZUrMrFq7o05bHRvSgbfNADNC2eSCPjehRXHeiThlwKxTkwM9vVn+MwgJI+dp9QkddKdApInIc1YyopqDwcPIPHSI/Kwvf4GBvhyMiItJgGMMVwCfWUljNIQYAm6y1Ke7xzLvA5cC6En26AXcXvf4K+OjY55t+QDjwBRBbzRikPOs+huWvwTl/g85Dsdby7KINPP/lJoZ2DeOFa/sS6Oc8+TjVcEWftnUz+XC88O7QYRAkvQbn3FW8cqRKti+H3Aw4bWjNxyciUkO0MqKaAiPcmXzVjRAREalxc4DtxvCEMXSpxv1tgW0l3qcVtZW0Cjh6ZMGVQFNjTKgxxgE8DdxTjc+VEzmQCh/fBW36wgUPUeiyPPDhrzz/5SZGxkYx4/p+tZaIqHcGjIGMrbD+8+rdv2kJGAd0PL9m4xIRqUFaGVFNwUXJiCO7dhHSsaOXoxEREWlQIoDrgJuAe4zhB+A1YK61ZNXQZ9wDvGiMGY17O8Z2oBD4K/CZtTbNnGR5uzFmDDAGIDw8nMTERAAyMzOLXwuE7fqajilv4p+7FzBsCuzN5q+/Y+bqXFbsKiS+oy8Xh+7n26Un3xXTWObWuII5y78VRxY+wapdTap8f9+VH2KbdmblT6sq1b+xzKs3aG5rh+a19nhybpWMqKag8HAA1Y0QERGpYdZyGJgJzDSG7sDNuGs6PGcM/we8Zi0/nmCI7UC7Eu+jitpKfIbdQdHKCGNME+BP1tqDxpizgUHGmL8CTQA/Y0ymtbZMEUxrbQKQABAbG2uPHjlZmeMnG43Vc+G7V9xFKwGwnJb6Dp/tbcWKXb2ZdGk3bjo3ptLDNaq59b2DgCWTiesWDmFdK39f1j5I3ARx91d6rhrVvHqY5rZ2aF5rjyfnVsmIagosSkZkKxkhIiJSa6xlrTE8C2QB9wFXA6ON4WfgVmtZXc5tSUBnY0wM7iTENbhXWhQzxrQC9ltrXcD9wCz359lRJfqMBmLLS0Q0CqvnwpIpkJEGIVEwZCL0HFm2n8sFmTvh4Nayjy1LwVVQqrspOMJVB2cRc+1SLuvVxkNfph7q+2dIfByWJUD8s5W/L+UrwMJpOtJTROo2JSOqyennR0BoqJIRIiIitcAYfHHXcrgZGAL8BIwF/g9oAfy76HWZPxlbawuMMXcCCwEnMMtau9YYMwVYbq2dD8QBjxljLO5tGnfU+peqT1bPpeDju/ApzHG/z9hG4Ud34Nz6AzSNhIOpcHCbO+GQkQau/NL3B7eG5tFlEhFHtTH7aKtExIkFh0KPq2DVuzBkkvvYz8pI/hICW0CbPrUbn4jIKVIy4hQERUSogKWIiEgNM4YXgGsBC8wG7ra21EkYR4xhPLCjojGstZ8Bnx3XNrHE63nAvBPFYa19A3ijiuE3CNmfTyToaCKiiNOVB8tnud80CXcnG9r2hW6Xu183b+9+DokCvyD3OE+cQdCR9DLjHwmMIKjWv0UDMGAM/DIHfnkbzv7ryftb6y5e2TGueqdwiIh4kJIRpyAoPJzD27advKOIiIhURTfgTuADa8mroM9eYLDnQmpcAo6Uv/LTZWFyzyVku3zJK3SRl+MiL9NF3lYXuQUu8gr2kFewy32twEXfQ1fyb59XCTLH/jVmWz+m5V/Nwx76LvVam97Q7ixI+g+cNRYcJzkIb9da95YZHekpIvWAkhGnICgigl3Ll3s7DBERkQbFWk662d1aCoCvPRBOo7TDFUqUY2/ZdtuKBesO4OfjcD+cjlKvmwb44F/i/Ue/nIfLwn0+c2lj9rHDhjKtYCSf5A5QMqKyBoyB92+BTYuhy4Un7pu8xP3c6YLaj0tE5BQpGXEKgsLDyT90iPysLHyDg70djoiISINgDFOBbdYy47j2sUBba3nIO5E1Hq/6Xc+/8l8i0ByrBZFt/XjV73pWPDis0uMkbTnA/IPnMT/vvFLtbZsH1lisDV7Xy9zbYpbNPHkyYtMSCOsGzVSPQ0TqvpOs9ZITCYqIAFDdCBERkZp1A7CynPYVwI0ejqVR6n3JGOYUuheouCykuVox0Y6h9yVjqjTOvcNPJ9C3dO2CQF8n9w4/vcZibfB8/CD2ZvfKiH3JFffLy4KtP+gUDRGpN5SMOAVHkxFHlIwQERGpSWHAnnLa9wHhHo6lUYoODSIfX/Jx0jX3Da4O+g/nXflXrujTtkrjXNGnLY+N6EHb5oEY3CsiHhvRo8rjNHr9bgKHLyz7T8V9tnwLhXnQSckIEakftE3jFASFu38eytLxniIiIjVpKzAISDmu/Q9AmufDaXyeXbSBv/uk4Ijsyfrbrjylsa7o01bJh1PVNBy6X+E+WeOCB8G/Sdk+m5aATyBEn+35+EREqsGjKyOMMRcZY9YbYzYZY8aXc32sMWaNMeYXY8y3xphunoyvqgKLkhHZSkaIiIjUpJnAs8ZwqzF0KnqMAZ4GErwcW4OXtGU/32/cRS9nCs52A7wdjhw14DbIPQSr3in/evIS6HAe+AZ4Ni4RkWryWDLCGOMEXgIuxn1k17XlJBvettb2sNb2BqYBz3gqvupw+vkREBqqZISIiEgNspancSckngc2FD2eA/5jLdO8GVtj8Mz/NjAgeDe+hUcgqr+3w5GjomKhTR/3Vg1rS187sAX2bdKRniJSr3hyZcQAYJO1NsVamwe8C1xesoO19lCJt8HAcf+nrXuCIiJUwFJERKSGWcv9QCtgYNGjtbWUWVUpNev75L38kLKPOzsfcDdExXo3IDnGGPcxn3vXQ0pi6Wubio70VPFKEalHPJmMaAtsK/E+raitFGPMHcaYZNwrI/7modiqLSg8XCsjREREaoG1ZFlLUtEj09vxNHTWWp5dtIHwZv6c5ZcMQa2gRQdvhyUldR8BQaFlC1kmfwkh0RB6mnfiEhGphjpXwNJa+xLwkjHmOuBB4M/H9zHGjAHGAISHh5OYmFh8LTMzs9T72pZRWEh2WppHP9MbPD2vjYXmtfZobmuH5rX2aG5LM4bBwLVANOBX8pq1XOCVoBq4bzftJWnLAR65vDs+K1a4t2gY4+2wpCTfAOg3Gr59Fg6kQov2UJgPKV9Dj6v070tE6hVPJiO2A+1KvI8qaqvIu8Ar5V2w1iZQVMAqNjbWxsXFFV9LTEyk5Pvatm7TJn75+mvO7d8f3+Bgj32up3l6XhsLzWvt0dzWDs1r7dHcHmMMo4EZwIdAHPAx0AWIAd7yWmANmLWWZxZtoE1IACPPbAoLN0DPq70dlpQn9hb4djokvQoXPgLblkHeYW3REJF6x5PbNJKAzsaYGGOMH3ANML9kB2NM5xJvLwE2ejC+agmKiABQ3QgREZGacw9wp7VcC+QD91tLH9yJCG3XqAWJ6/ewcutB7rygM/67VrobVS+ibgppC13j4ec3IS/bfYqGcULMH7wdmYhIlVQtGWFMa4xpXeJ9D4x5FGOuPdmt1toC4E5gIfAbMNdau9YYM8UYc1lRtzuNMWuNMb8Ad1POFo26pjgZoboRIiIiNaUjsLjodS7QpOj1i8BobwTUkB1dFdGuZSD/LzYK0lYABtr09XZoUpEBYyDnIKx5z128st0ACAjxdlQiIlVS1W0ac4HZwCyMaQV8A+wA7sKYNlj79IluttZ+Bnx2XNvEEq//XsV4vE4rI0RERGrcPqBp0evtwJnAaiAUCPRWUA3V4t92s2Z7BtOu6omv0wFpSRDWFQKaeTs0qUj7c6FpW/j0bnAVgH8IrJ4LPUd6OzIRkUqr6jaNnsCPRa+vAjZhbXfgRuC2mgysvggMCwO0MkJERKQGLQUuLHo9F3jeGF4H3gEWeS2qBsjlcq+KaB8axIg+bcFadzJCWzTqtjXvQfZudyICIDcDPvmbOyEhIlJPVDUZEcixvZpDOVbz4WdKF6dsNJx+fgSEhioZISIiUnPuxJ14AHgMeBL3qoi5wF+8FVRDtHDtTn5LP8Tfh3TGx+mAfcnu5f9R/b0dmpzIkinuUzRKyj/ibhcRqSequk1jIzACY97H/ReLJ4vaw4GDNRlYfRIUEaFtGiIiIjXAGHxwF7n+CMBaXMATXg2qgXK5LM8u3kDH1sFc3rutuzEtyf2sZETdlpFWtXYRkTqoqisjJuP+gWAL8CPW/lTUPhxYWYNx1StB4eFaGSEiIlIDrKUA9x87fL0dS0P36Zp0NuzKZNzQLjgdxt2YlgT+zaDV6d4NTk4sJKpq7SIidVDVkhHWfgBEA7HARSWuLMZ9+kWjpJURIiIiNepHoJ+3g2jICl2W6Ys30CW8CfE9Io9dSEuCtn3B4cnT36XKhkwE3+NqufoGuttFROqJqm7TAGt3Acd+8zbmNGAV1ubUXFj1S1B4OPmHDpGflYVvcLC3wxEREanv/gM8ZQzRwAogq+RFa/nZK1E1IPNXbSd5TxavjOqL4+iqiLxs2LUWzvuHd4OTkzt6asaSKe6tGSFR7kSETtMQkXqkaskIY/4NrMfa/2KMAf4HDAEyMOaiEts2GpWSx3uGdOzo5WhERETqvbeLnp8p55oFnB6MpcEpKHTx3OKNdI1sxvDuEccupP8CtlD1IuqLniOVfBCReq2qKyNGAVcXvb4Y6A0MLGp/HBhcc6HVH8XJiJ07lYwQERE5dTHeDqAh+2Dldrbsyybhhn7HVkVAieKVOtZTRERqX1WTEeHA0TK9fwTmYu0yjNkPLK/RyOqRkisjRERE5NRYS6q3Y2io8gtdPL9kIz3ahjCsW3jpi2lJ0CIGglt5JzgREWlUqpqM2Ae0x52QuBAYX2IcU9FNDV1gWBgA2enpXo5ERESk/jOGESe6bi0feCqWhmbeijTSDhzhkcvPLNpxW8Ra2JYEMX/wXnAiItKoVDUZ8T7wNsZsAFoCC4vaewObajKw+sTp50dAaKhWRoiIiNSMeRW026Jn1YyohtyCQl5YspHe7ZoTd3rr0hcPbYfMnaoXISIiHlPVc5vuBp4H1gHDsPZodetI4JWaDKy+0fGeIiIiNcNaHCUfgB9wFrAU0J/uq2lu0jZ2ZORw97AupVdFgOpFiIiIx1VtZYS1BcDT5bQ/W0Px1FtB4eEc3rbN22GIiIg0ONZSACQZwwO4//jRy8sh1Ts5+YW8+NUm+ndowaDO5dSESFsOPgEQfqbngxMRkUapqts0wJhw4A6gG+7lkuuAl7B2d82GVr8ERUSwa3mjreEpIiLiCQeBTt4Ooj56+6et7DqUy7NX9y67KgLcKyMie4GPn+eDExGRRqlq2zSMORd3bYjrgCNADu5jPTdhzNk1Hl09EhQeTv6hQ+RnZZ28s4iIiFTIGPoe9+hnDPHATGClt+Orb47kFfJyYjIDO7bknE7lrIooyIP0VaoXISIiHlXVmhFPAe8AXbD2Bqy9AegCvEt52zcaER3vKSIiUmOWA0lFz0dfz8dduPIvlRnAGHORMWa9MWaTMWZ8OdfbG2OWGGNWG2MSjTFRRe29jTE/GGPWFl27usa+lZe89WMqezNzuXvY6eV32PUrFOSoXoSIiHhUVbdp9AZGY62ruMVaF8Y8QyP/S0VxMmLnTkI6dvRyNCIiIvVazHHvXcAea8mpzM3GGCfwEjAM93HkScaY+dbadSW6PQW8aa39rzHmAuAx4AYgG7jRWrvRGNMGWGGMWWitPXiK38krsnILeOXrZAZ1bsWAmJbld0or2maqlREiIuJBVU1GZOD+AWH9ce0xuPdxNlpaGSEiIlIzrCX1FIcYAGyy1qYAGGPeBS7HXefqqG64TwkD+Ar4yP3ZdsOxOOwOY8xuoDX17Oecj1Zu58mF69l+8AgAfaNbVNw5LQmaRkKzth6KTkREpOrbNN4FXsOYURgTU/S4HngV9/aNRiswLAyA7PR0L0ciIiJSvxnDVGMYW077WGN4pBJDtAVKHnGVVtRW0ipgRNHrK4GmxpjQ0p9nBuA+VjS5srHXBR+t3M79H6wpTkQAJHyTwkcrt5d/Q1qSe4tGeYUtRUREaklVV0bcBxhgVtG9BsjDfcxWmf2YjYnTz4+A0FCtjBARETl1NwD/r5z2FcD9wEM18Bn3AC8aY0YD3wDbgcKjF40xkcBs4M+25PbUEowxY4AxAOHh4SQmJgKQmZlZ/NobHknM5ki+LdV2JL+QRz5eRfOMjaXaffMyOPfAZpJb/IFtXoy5srw9tw2V5rX2aG5rh+a19nhybquWjLA2D/g7xtzPsaO1krE2u6YDq4+CIiKYn7GUq6d3YGvGVqJDopk6ZCqjeozydmgiIiL1SRiwp5z2fUB4Je7fDrQr8T6qqK2YtXYHRSsjjDFNgD8drQthjGkGfApMsNb+WNGHWGsTgASA2NhYGxcXB0BiYiJHX3vD/i8+Lb89x5aNa/0X8D10Ov8aOrU/p/aDO0XentuGSvNaezS3tUPzWns8ObcnT0YYM78SfdzP1l52qgHVZ99FZ/B00DfkZrj/sJKakcqYT8YAKCEhIiJSeVuBQUDKce1/wL3l4mSSgM7GmBjcSYhrcB9LXswY0wrYX7Tq4X7cqz4xxvgBH+IubjnvVL6Et7RpHlhqi0bJ9jLSksA4IbK3ByITERE5pjI1I/ZV4dGovdr0J3KdhaXasvOzmbBkgpciEhERqZdmAs8aw63G0KnoMQb3MeIJJ7vZWlsA3AksBH4D5lpr1xpjphhjjv7hJA5Yb4zZgHu1xdSi9pG4kx6jjTG/FD3q1W/q9w4/HX+f0j/iBfo6uXd4OUd7bl8O4d3BL8hD0YmIiLidfGWEtTd5II4GYRcZ5bZvzdjq4UhERETqL2t52hhaAc/jLiAJ7hpVz1nLtMqNYT8DPjuubWKJ1/OAMisfrLVvAW9VM/Q64Yo+bVm3I4OEpZsxuFdE3Dv8dK7oc1wNT1chpK2AniO9EqeIiDRuVS1gKRVYlLwIg8Fiy1yLDon2QkQiIiL1l7XcbwyP4j6CE+A3a8n0Zkz1SVizAAB+fmgYLYL9yu+0dwPkHYao/h6MTERExK2qR3vKcfIK87hv0X1c+NaFhPmH4ltY+lisAGcAU4dMreBuEREROZ4xRBhDlLVkWUtS0SPTGKKMqVQBy0YveU8mLYP9Kk5EgLteBCgZISIiXqFkxCnYtH8T5846lye/f5Lb+t3Gqmu+59ZlkbTxbY3B4DAOTmt5mopXioiIVM1bwMXltA/HfdymnETy7iw6tQ4+cae0JAhoDqGdTtxPRESkFigZUU1zVs+hz8w+bNq/ifdHvs+M+BmEtmnPudua80Xzh3FNcjFt6DR+3fMrS1OXejtcERGR+iQW+Kac9qVF1xq21XPh2TPh4ebu59VzqzxE8p5MOrVucuJOacvdqyKMOXE/ERGRWqBkRBUdzj3MjR/eyPUfXk/viN6sGruKEV1HAOD08yMgNJTsXbsAuL3/7UQ0iWBi4sQTDSkiIiKl+QD+5bQHVNDecKyeC5/8DTK2Adb9/MnfqpSQOJidx76sPDqeaGVEziHY/RtENfzcjoiI1E1KRlTB8h3L6ZvQlzlr5vDw+Q/z1Z+/KlOcMigiguydO92vfYO4/7z7SdySyFebv/JGyCIiIvXRT8Dt5bTfASR5OBbPWjIF8o+Ubss/4m6vpOQ9WQAnXhmx42fAKhkhIiJeo2REJbisi6e/f5pzXjuH3IJcEv+cyKS4Sfg4yh5GEhQRUbwyAmBMvzG0adqGiYkTsbbsSRsiIiJSxgTgz8bwnTE8UvT4DrgReMDLsdWujLSqtZcjeY/70JETJiPSlruf2/ar9LgiIiI1ScmIk9iZuZOL51zMPYvuIb5LPL+M/YVB7QdV2D8oPLx4ZQRAgE8AEwZN4Nut37IoZZEnQhYREanXrOVH4GxgCzCi6JECDASCvBeZB4REVa29HMl7MvFzOohqEVhxp7Tl0KoLBLaoYoAiIiI1Q8mIE1i4aSG9ZvTim9RvmHHJDN4f+T4tA1ue8J6giAjyDx8mPyuruO2WPrcQHRLNxK+0OkJERKQyrGWVtYyylu64T9HYAHwILPRuZLVsyETwPS6J4Bvobq+k5N1ZdGgVhI+zgh/zrHWfpKEjPUVExIuUjChhzpo5dJjeAcdkByGPhXDRnIsICw5j+a3LuS32Nkwlqk0HhbuPPy+5OsLfx58HBz3IT9t/4rONn9Va/CIiIg2FMTiNYYQxfApsBq4AZgCneTeyWtZzJFz6PIS0A4p+7ug+wt1eSSl7MunY6gRbNA5sgey9qhchIiJepWREkTlr5jDmkzGkZqRisRzKO4SPw4e7B95N97DulR4nKCICoFTdCIDRvUcT0zxGtSNEREROwBhON4YngR3AU8BK3L+V32At06xls1cD9ISeI+Efv8KkA9DuLEhJhILcSt2aX+hi6/5sOoWd4CSNo/UitDJCRES8SMmIIhOWTCA7P7tUW4GrgMlfT67SOBUlI3ydvkw8fyI/p//M/PXzTy1YERGRBsgYlgI/Ai2AkdbS0VoeBBpnFt8YiLsfDm2Hn9+s1C2p+7IpcNmTFK9MAt9gaN21hgIVERGpOiUjimzN2Fql9ooEhoUBkJ2eXuba9T2vp3PLzkxMnIjLuqoepIiISMN2NvAm8Ky1fO3tYOqEjnEQfTYsfQbyc07avXInaSRBmz7gLHsqmIiIiKcoGVEkOiS6Su0V2fq//4HDwZqXXuKjoUPZvGBB8TUfhw+Tzp/E6l2r+eC3D04pXhERkQaoP+ADfGsMK43hH8YQ4e2gvOro6ojDOyq1OuJoMqJj6wq2aeTnwM41qhchIiJep2REkalDphLkW/q0sCDfIKYOmVrpMTYvWMCySZPA5V71kJ2ezrJJk0olJK458xq6turKpMRJFLoKayZ4ERGRBsBaVlrLHUAk8AxwGbAN988rlxhD4zyHMuYP0P5cWPo05B85Ydfk3VmEN/OnaYBv+R12rgZXvupFiIiI1ykZUWRUj1EkXJpA+5D2GAztQ9qTcGkCo3qMqvQYq6ZPpzCn9BLKwpwcVk2fXvze6XDycNzDrNuzjrlr59ZY/CIiIg2FteRYy2xrGQx0BZ4E/gHsNIbPvRudFxgDgx+AzJ2w4o0Tdk3Ze5KTNNKS3M9aGSEiIl6mZEQJo3qMYsu4LbgmudgybkuVEhFQ+jjPE7Vf1e0qzgw7k4e/fpgCV0G14xUREWnorGWTtYwH2gEjgTwvh+QdHc6DDoPg22crXB1hrSV5d+ZJTtJIgpBoaNq4d7+IiIj3KRlRg46epHGydodxMDluMhv2beDtNW97IjQREZF6zVoKreVja7nc27F4zeAHIHMXLJ9V7uWz+/XcAAAgAElEQVS9mXkcyik4SfHK5VoVISIidYJHkxHGmIuMMeuNMZuMMePLuX63MWadMWa1MWaJMaa9J+M7Vb3GjcMZEFCm3b9lS1z5+aXarjzjSvpE9GHK11PIL8wvc4/UXXPWzKHD9A44JjvoML0Dc9bM8XZIIiLSGLQ/B2LOd6+OyMsqc/mkJ2kcSoeMbaoXISIidYLHkhHGGCfwEnAx0A241hjT7bhuK4FYa21PYB4wzVPx1YSY+HgGTJ5MUGQkGENQZCQdLruMA2vX8v2//oWr4NiWDGMMk+Mmk3wgmTdXVe7scPG+OWvmMOaTMaRmpGKxpGakMuaTMUpIiIiIZwx+ALL2QNJrZS4VJyPCKkhGbF/uflYyQkRE6gBPHjA9ANhkrU0BMMa8C1wOrDvawVr7VYn+PwLXezC+GhETH09MfHypthann87KJ5/E+Phw9mOP4XA6AYjvEk//Nv155JtHuKHXDfg5/bwRslTBhCUTyM7PLtWWnZ/NhCUTqlxjREREpMqiB0LHwfDdc9D/FvA7Vh8ieXcWAb4OIpuVXaUJuOtFOHwhooeHghUREamYJ7dptMV9PNdRaUVtFbkFGkbF7K6jR9Nr3DhSP/2Unx58EFeh+0hPYwxTBk8hNSOV11e+7uUopTK2ZmytUruIiEiNG/wAZO+FZf8p1Xz0JA2Hw5R/X9pyiOwJvhUkK0RERDzIkysjKs0Ycz0QC5xfwfUxwBiA8PBwEhMTi69lZmaWel9ndO5M0/h4Ns+fz649ewi57jqMw4G/9ad7s+48tPghYg7F4Oeom6sj6uy8eliYfxi7cneV216d+dG81h7Nbe3QvNYeza1UWrsBcNrQY6sj/JsC7m0avdu1KP+ewgLYsRL63ujBQEVERCrmyWTEdtzHch0VVdRWijFmKDABON9am1veQNbaBCABIDY21sbFxRVfS0xMpOT7OiUujtVRUfw6YwZt2rWj/8SJGGN4rv1zDJ09lA1NNnDngDu9HWW56vS8etCjzR7l1k9uLdUW5BvE05c8TVyPuCqPp3mtPZrb2qF5rT2aW6mSuAfg1QtgWQIM+ic5+YWkHTjCn/pGld9/9zrIz1a9CBERqTM8uU0jCehsjIkxxvgB1wDzS3YwxvQBZgKXWWt3ezA2j+lx5510+8tf2DR3LsunTsVaywUxF/CH9n/g30v/zZEKzg6XuiF5fzIAEU0iMLiXwY7pO0b1IkRExLOi+kHnC+H7FyDnEJv3ZmHtCU7SSEsquk/HeoqISN3gsWSEtbYAuBNYCPwGzLXWrjXGTDHGXFbU7UmgCfCeMeYXY8z8Coart4wx9Bo3jjNGj2bjO+/w8xNPADAlbgrpmenMWD7DyxFKRVIPpvLsj89yQ88bSP9nOvkP5dOuWTvW7lnr7dBERKQxihsPRw7AspknP9YzbTkEt4bm9erUdBERacA8WjPCWvsZ8NlxbRNLvB7qyXi8xRhDn3vuwRYUsH72bBw+Pvzhn/9kSMwQHv/uccb0G0NwierYUjfcv+R+jDFMvWAqAE6Hk1v73srExIls3LeRzqGdvRyhiIg0Km37QZeL4fsXSevj/hEqplUFPz+kJbm3aJgKiluKiIh4mCe3aUgJxhj6jh9P52uu4bfXX2f1888zOW4yu7N281LSS94OT47zY9qPvPPrO9xz9j20CzlW+uQvff+Cj8OHhBUJXoxOREQarbjxkHOQdhvfpG3zQAL9nGX7ZO+HfRu1RUNEROoUJSO8yBhD7IQJnPb//h9rExJo+skvXHTaRUz7bhqHcw97OzwpYq3l7oV3E9Ekgn+d969S1yKbRnLFGVfw+i+vk1OQ46UIRUSk0WrTG06/hPP3/R89WlXQZ/vP7ue2SkaIiEjdoWSElxmHg/4TJ9JxxAh+ffllbtzfi31H9vH8T897OzQp8t669/gh7QceHfwoTfzK7sUd228s+47s472173khOhERKY8x5iJjzHpjzCZjzPhyrrc3xiwxxqw2xiQaY6JKXPuzMWZj0ePPno286lzn/4smNotrCj8pv0NaEmCgbV+PxiUiInIiSkbUAcbhYMDDD9PhssuwMz6hq087HvrqIRyTHXSY3oE5a+Z4O8RGK6cgh38t/hc9w3syuvfocvtcEHMBXUK7MGOFio+KiNQFxhgn8BJwMdANuNYY0+24bk8Bb1prewJTgMeK7m0JTALOAgYAk4wxLTwVe3XsDOrCF4X9OWfPXHdBy+NtXw5h3cC/qeeDExERqYCSEXWEw+lk4KOPsv6yDmzKTcMW/ZOakcqtH96ihISXvPDTC2w5uIWnL3wap6Ocfbi4t9uM7TeW77d9z+pdqz0coYiIlGMAsMlam2KtzQPeBS4/rk834Mui11+VuD4cWGSt3W+tPQAsAi7yQMzVlrwnk+kFf8KvIBN+eLn0RZfLfZKG6kWIiEgdo2REHeJwOnm1aRL5Tluq/YjN5b4F//BSVI3Xnqw9PLr0US7pfAlDO574oJc/9/4zAT4BvJL0ioeiExGRE2gLbCvxPq2oraRVwIii11cCTY0xoZW8t05J2ZPF7zaanM6Xwo+vuAtWHrU/GXIOuk/SEBERqUM8erSnnFx63h4o59St9Nw9rHj8cZp17EhIx440i4nBv2VLzAmO6Nq8YAGrpk8ne+dOgiIi6DVuHDHx8bUYfcPycOLDZOVl8eSwJ0/at2VgS67ufjVvrXmLacOm0VRLYUVE6rp7gBeNMaOBb4DtQGFVBjDGjAHGAISHh5OYmAhAZmZm8WtP+GZdLoE+sDpkKP3zFrD13XvZ3PEGAMJ3fklXYFm6JduDMdUWT89tY6F5rT2a29qhea09npxbJSPqmNBsX/YG55dp9yswrPtgLs6s3GNtzZrRrGNH9yMmxv3o2JEmbduS+sUXLJs0icIc9wkP2enpLJs0CUAJiUpYt2cdM1fMZGzsWLq27lqpe26PvZ3/rvovc9bMYWzs2FqOUERETmA70K7E+6iitmLW2h0UrYwwxjQB/mStPWiM2Q7EHXdvYnkfYq1NABIAYmNjbVyc+7bExESOvvaE/2z6kS6RhQyIvwSOJNJ+4xe0v3oaBIfCgo/BvxkDLr4eHPV/Qayn57ax0LzWHs1t7dC81h5Pzq2SEXXM9ds68/Jpv5Hnc2yrhtMFub6Wp24q5L/nzyBsv+HQ5s3uR0oKO77+mpQPPiju7/D1xVqLLSgoNXZhTg6rpk9XMqIS7l10L038mjDp/EmVvmdA2wH0jujNK8tf4bZ+t51w1YqIiNSqJKCzMSYGdxLiGuC6kh2MMa2A/dZaF3A/MKvo0kLg3yWKVl5YdL3OSt6dxTmnhbrfnD8e1n4EP7wAQx92n6TRtl+DSESIiEjDomREHfO3UU9QkHAn756xnb1BBbTK9uGa39vS7+qbuGf7C/zh40t47bLX+H/n3VDqvryMDA5t2UJGSgqHUlL4bdascsfP3rnTE1+jXluUvIjPNn7GtKHTaB3cutL3GWO4PfZ2bltwGz+m/cjZ7c6uxShFRKQi1toCY8yduBMLTmCWtXatMWYKsNxaOx/36ofHjDEW9zaNO4ru3W+MeQR3QgNgirV2f5kPqSMycwvYeSiHTq2Ljp4OOwPOHAE/JUC/m2DXWhj0T+8GKSIiUg4lI+qYmPh47gaGlFPrYUjGzYycN5KR80Zy19a7eOrCp/Bz+gHgFxJCq169aNWrFwCpn39Odnp6mfF9AgLI3L6dJm3rdC0uryl0FfLP//2TmOYx3HXWXVW+/7oe13HP/+7hleWvKBkhIuJF1trPgM+Oa5tY4vU8YF4F987i2EqJOi1lTyYAnVoHH2s8fzz8+j68GAvWBctfh1ZdoOdIL0UpIiJSltbs1UEx8fFcsXgx1/36K1csXly8raJdSDu+Hv01/xj4D15Y9gKDXh9E6sHUcsfoNW4czoCAUm3G6aQgL48Ff/wjSVOmkL1rV61/l/rm9V9eZ83uNTwx9AkCfAJOfsNxmvg14YaeNzB37Vz2Ze+rhQhFRESOSdmTBXBsZQRA+i9gnFCY536fvRc++RusnuuFCEVERMqnZEQ94+f045nhz/D+yPf5fe/v9JnZh083fFqmX0x8PAMmTyYoMhKMISgykoH//jdX/O9/dPrTn0h+/33mX3QRKx5/nCN793rhm9Q9h3MP8+CXD3JOu3O4qttV1R7n9v63k1uYyxu/vFFzwYmIiJQjeU8mTochOjToWOOSKWCPOxgk/4i7XUREpI5QMqKeGtF1BD+P+ZkOzTsQ/0489y++nwJX6YKV5a2wCIqIoP/EicR/+ikdLrmEDW+/zfyLLuKXZ54h9+BBL32buuGJ755gV9YunrnwmVMqPnlm2JmcF30eM1bMwGVdNRihiIhIacl7MoluGYS/j/NYY0Za+Z0rahcREfECJSPqsU4tO/H9Ld9zW7/bePy7xxny5hB2HN5RqXubREUx8NFHuWT+fKIuuIB1s2bx8YUXsvrFF8k7fLiWI697tmVs4+kfnubaM6/lrKizTnm8sf3Gsmn/Jr7c/GUNRCciIlK+5N1ZpetFAIREld+5onYREREvUDKingvwCWBG/AxmXzmb5TuW02dmnyr9AtysQwfOnTaNP370EZHnnMOvr7zC/AsvZG1CAvlZWWxesICPhg5lxx138NHQoWxesKDKMR4d4+0zz6z2GLXtgS8fwFrLY0Meq5Hxrup2Fa2CWvHK8ldqZDwREZHjFbosm/dl0bFkvQiAIRPBN7B0m2+gu11ERKSOUDKigbi+5/Uk3ZpEaGAow2YP45GvH+Gt1W/RYXoHHJMddJjegTlr5lR4f/PTTmPQ9OlcNG8erfr0YdVzz/HB4MH8OGFC8akc2enpLJs0qUrJhM0LFrBs0iT3GNZWa4zalrQ9ibdWv8XdZ99N++bta2RMfx9/bu59Mx///jHbD22vkTFFRERK2n7gCHkFrrIrI3qOhEufh5B2gHE/X/q8TtMQEZE6RUd7NiDdWndj2a3LGLtgLBMTJ+IwjuKaBakZqYz5ZAwAo3qMqnCMll27Evfyy+xdtYrFN92ELShdh6IwJ4ekKVM4uH49xunEGON+djiOPZd4/esrr/BN653M7bmbvUEFtMr2YeTqMPynTy8+JcSbrLXc/b+7CQsOY/x542t07DH9xjDt+2m8tvI1Jp6vv0aJiEjNSi4+1rNJ2Ys9Ryr5ICIidZqSEQ1ME78mzL5yNp9v+pz9R/aXupadn82EJRNOmIw4qlWvXrjy8sq9VpCVxYY5c7CFhViXC+uquEjjd+0O8mr/dPJ8LAB7gwt4tX86LDdcUYXvVVs++O0Dvt36LTPjZ9LMv1mNjt2pZSeGdxpOwooEHhj0AD4O/ecmIiI154TJCBERkTpOvx01QMYYDhw5UO61rRlbKz1OUERE8RaNUu2RkVyxeHHxe2stWHssOVFYiC16P+6xsOJExFF5Ppa5vfbyUqUjqR25Bbnct/g+urfuzs19bq6VzxgbO5Yr/+9KFmxYwBVn1IX0i4iINBTJezJpGexHi2A/b4ciIiJSZaoZ0UBFh0SX2x7oG0jKgZRKjdFr3DicAQGl2pwBAfQaN65UmzEG43Dg8PXF6e+PT1AQvsHB+DVrxt6A3HLH3huQy+/vvuNOZHjJS0kvkXIghacvfLrWVi3Ed4knqlkUM5bPqJXxRUSk8Sr3JA0REZF6QsmIBmrqkKkE+QaVavN1+JJfmM8ZL57B3z//O3uz955wjJj4eAZMnkxQZCTgXhExYPLkStV6WLFjBVfNvYoKUw0Gzll5I6MeHMCKLT9V5ivVqL3Ze5ny9RQuOu0ihp82vNY+x8fhw619b2Vh8kKS9yfX2ueIiEjjk7wnk46ttEVDRETqJyUjGqhRPUaRcGkC7UPaYzC0D2nP61e8zpZxWxjdezQvJr1Ip+c78djSx8jOz65wnJj4eK5YvJg2L73EFYsXnzARYa0lcUsiw98aTux/YlmcspjLulxGoE/p48WCfIK4s/+d9GrRjbk+y4n970DOfO4Mnvr+KXYc3lFjc1CeOWvm0GF6B1o/2ZqM3Azi2sfV6ucB3NLnFpzGycwVM2v9s0REpHE4mJ3Hvqw8OoVpZYSIiNRPSkY0YKN6jGLLuC24JrnYMm4Lo3qMok3TNiRcmsCa29cQ1yGOB758gC4vdGHWylkUugqr9Tku6+KT9Z9wzqxzGPzfwfyy8xceG/IYqeNS+fjaj/nPZf8plRRJuCyBF/74Al+NX8WqYQv5y4bTyNm8jXsX3Uu7Z9tx4ewLmb1qNpl5mTU6H3PWzOHWD28hNSO1uG3yl5NOeORpTWjbrC2Xn3E5s1bOIreg/G0rIiIiVZG8JwtQ8UoREam/lIxopLq17sbH13zMN6O/IapZFLfMv4VeM3rx6YZPK13HocBVwJzVc+g1oxeXvXsZOzN38tIfX2LL37cw/rzxhASEAOUnRY7qft6FPP/cDyRkXsuTn3fihvyBbNy3gRs/upGIpyK48cMbWZS8iEJXYfGqBsdkBx2mdzhhEiErL4vk/cl8t/U75q2bx4vLXmTsR2M4YksnA47YXO5b8I9qzGDV3B57O/uO7GPeunm1/lkiItLw6SQNERGp73SaRiM3qP0gfrjlB97/7X3uX3I/8e/Ec37783ly2JP0b9u/3HtyCnJ445c3mPbdNDYf3Ey31t1484o3uebMa/B1+lY5hsBWrYibOZPWM2fS5uWX+VPHHviMf4wP933F3LVzmb16Ns39m5OZn0mBqwCA1IxUbv74Zj7+/WMimkSwM3Nn8SM9M738VRUWMGWb03P3sG3xYpp36UKTqCiMo+ZzdBfEXMBpLU/jleWvMKrnyY9WFREROZHkPZn4OR1EtQg8eWcREZE6SMkIwRjDVd2u4vLTLydhRQKTv57MgFcHMLL7SAa2HchzPz3H1oytRK2M4tx255KYmsjOzJ0MaDuAZ4c/y6WnX4rDnNov8A6nkx5//Sut+/bl+/vuI/tvTzP+oYd4/p7nWbBhATd+eGNxIuKovMI83lv3HiH+IUQ0iSCiSQR9I/sS2SSSiCYRtHKG4Ny0k8Kk3yj48VcmxG1gX3BBmc8OzfZh6d//DoBPYCAhnTvTvEsXmnfpQovTT6d55874hYQU99+8YAGrpk8ne+dOgiIi6DVu3EmLejqMg7H9xnLPontYs2sNPcJ7nNJ8iYhI45a8O4v2oUH4OLXIVURE6iclI6SYr9OXOwbcwQ29buCp75/i8W8fZ+7aucXXtx3axrtr36V76+7MGTGHwR0GY0w5Sw1OQcTAgVz8/vt8d++9/PjAA3RasYIrHniAnIKccvsbDAfHHyx+n3f4MNu/+orUt79g53ff4SooILhNG6JH3sT1y17lldM3kOdzbBuKX4Hh+q1dGP7uuxzcsIEDGzZwcP16ti1aRPK8Y1sqgiIiaN6lCzgc7nHz8wHITk9n2aRJACdNSIzuPZoJX05gxvIZvHTJS9WeIxERkZS9mXQJa+rtMERERKpNyQgpo5l/M6YMnsKslbPYfnh7meuZeZlcEHNBrX1+YOvWXPDqq6x56SXWJiSwb80aos5tw7assrFEh0STn5lJWmIiW7/4gvRvv8WVn09QRARdrr+e6OHDCe3RA2MMzRecQWHCnbx7xnb2BhXQKtuHa35vy9/GPEFojx6E9ji2WsFay5E9ezi4fj0HN2wo9TheYU4Oq6ZPP2kyIjQolJHdRzJ79WyeGPYETfy0z1dERKouv9DF1n3ZXHxmhLdDERERqTYlI6RCFR2zuTVja61/tsPHh15//zut+/blh/HjuTTRwat9TJlVDSN/Def9QYNw5eW5ExDXXedOQPTsWWbVRkx8PHcDQyqxxcIYQ1BYGEFhYbQZNKi4/e0zz4RyCnxmp6fz44MPEj18OBEDB+LwLb92xu2xtzN79WzeXvM2Y/qNqebsiIhIY5a6L5sCl1XxShERqdeUjJAKRYdElzoGs2S7p7QZNIiL5s0j/+KLoaCQuT13F69qGLk6jD7bs+k8ahTRF11Eq549T1p8MiY+/qQrGE4kKCKC7PT0Mu3OwEC2LVpEyocf4hcSQruhQ2l/8cWE9e+Pw+fYf2YDowbSK7wXryx/hVv73lrtOEREpPHSSRoiItIQKBkhFZo6ZCpjPhlDdn52cVuQbxBTh0z1aBzBkZG4Cgo4d1tzzt3WvNQ1ayz9xo/3WCy9xo1j2aRJFOYcq2HhDAhgwMMPEz1sGOnffUfqF1+Q+vnnJL//Pv4tW9Ju2DDaDx9O69hYHE4nY2PHcvunt7Ns+zKPxS0iIg3H0WREx9bBXo5ERESk+lSCWSo0qscoEi5NoH1IewyG9iHtSbg0gVE9PH80ZVBE+ftiK2qvLTHx8QyYPJmgyEgwhqDISAZMnkxMfDxOf3+iLriAc6dNY8TSpQyaPp3ws85i8/z5LLn5Zj4aPJikRx/lwsJuBDsCuf/fV7Ljjjv4aOhQNi9Y4NHvUdKcNXPoML0DjskOOkzvwJw1c7wWS0OjuRWR2pC8O4uwpv40Daj6cdoiIiJ1hVZGyAmN6jGKUT1GkZiYSFxcnNfiqGhFQq9x4zweS2W2evgEBNBu2DDaDRtGQXY2O5YuJfXzz0n54AM2vvMOZ/cN4OsOO7nGrxlU4USOmjZnzRxu/fAWjthcAFIzUrn1w1sAvJJ0akjmrJlTamVRakYqYz5x1wnR3IrIqUjZm6ktGiIiUu9pZYTUCydakVDX+QQFET18OIOmT2fE0qX4N29Oyywf8n0st12+nr9fsoFvWu9k+aOPsufnnynMy/NYbPd98o/iRMRRR2wu9y34h8diaKgmLJlQaosTQHZ+NhOWTPBSRCLSEFhrSd6dSacwbdEQEZH6TSsjpN441eKTdYFvcDBfNkvlo+573Q0G9gYX8Gr/dEiC/BtuwOnvT6tevWgdG0t4bCyhPXviExhYI5/vsi5W7VzFl5u/5MstX7Ijbw+Ysv125O7hUO4hmvk3q5HPbWzyC/PLLf4KnjmNRkQarr2ZeRzKKdDKCBERqfeUjBDxsPd67yt1RClAno/l1QE78Rk5mMh0FxnrdpA+8xUcL1scPj6E9uhB69hYwmJjad2nD77B7r+IbV6wgFUnOKrU5XKxKuUnFq6Zz1epifx4aDWHrPuv9W0yAwjwd5Dj6yobpIHwp8K58owrubHXjQztOBQfR+3/72LOmjlMWDKBrRlbiQ6JZuqQqV7Z0nCyea2ItZZ56+bx4FcPVtgnqllUTYYqIo2MTtIQEZGGQskIEQ/bG5Bbbnue08WLae/gsi7oAv5d/ekcGE1MThARO9bS6tPviZrtS/N8P1p27UZAy5bM2/Yp/9c3vei40/WMnLGB8xa/xy+B6fyU+zu/BOwkwz8fgFZZvvTdHUyf/I4MbNKDmLZnMPuXN5nZPaVUcsSvwBC/vhW+/bryxcbPeefXdwgPDmdUj1Hc2OtGekX0qpV5qSv1KzYvWFCqPkl2JWt6LE5ZzPjF41mRvoLurbtz98C7mbFiRpmtGrmFuSxNXcqg9oNq70uISIOlkzRERKShUDJCxMOiQ6LLXcLfPqQ9v9/5O7/t+Y01u9ewetdq1uxew/Jda0gPS4cwd78WJpj22XtxHTzMur5ZFDjd7XuDC3g5dhsvmzcBCPUN4myfrpwbGsuQjkPpfsZAgtu2xennV/yZrRb0xSbcybtnbC9KaPhw9bo2/NE/loNv/k58YHt2XDqcpS0P8MKyF3jmx2foGd6TG3veyHU9riOyaSRQ/RUNuQW5pBxIYdP+Tfz149sqrF/hyWTEqunTSxVKBSjMyWHV9OnlJiOW71jO/UvuZ3HKYqJDonnj8je4vuf1OB1O+rbpW2peRvUYxTu/vsP5b5zPXQPu4t9D/k2wX/34haJ4tUh6Oh9FRlZ6tYiI1KyUPVkE+DpoE1Iz2/dERES8RckIEQ+bOmRqqVMWAIJ8g5g6ZCoBPgH0iexDn8g+pe7Zm72XNbvWlEpSLEtbVrbeg4HgXAfL717L6aGnY0w5BSFKiImP525gSDlbEjKSk/n9zTfx+Wg+1+bnc+vgG/nt/FA+PPg19yy6h/sW38ewjsPo1KITr//yOkcKjgBlT43IK8xj84HNbNy/kY37NrJp/yb36/0b2Zqx1b0S5ATSc/dUbmJPkbWWA7//TnZ6ernXs9PT2ZWUROvevXH4+rJh3wYe/PJB3lv3HqGBoTw7/FnGxo4lwCeg+J6jp9GUdP+g+3lgyQM8v+x5FmxcwGuXvUZch7ja/GqnrLqrRUSk5iXvyaRjqyY4HCf+/7uIiEhdp2SEiIcd/eW0KisJWgW1YnDMYAbHDC5uczxssOX0zfZzcUarMyodT0WFQUM6deKsyZPpedddbHz3XTa+8w7tvjzIw2eeid/Vt/Blsy289escFiYvLBtDfjZ/mf8XHvryIVIzUkslHJoHNKdzy86cFR7LlS0H03qvJWTzIR4KnM/+oIIyYwXlOcjLOYJfQO38FTAzLY0tn37KlgULOJSScsK+S0aP5nDrAD47t4BP/X4lwDeQiX+YyD/P+Weli3028WvC8xc/z1XdruLmj29m8H8H89fYv/L40Mdp6t+0Jr5SjavqahERqT3JezLp3a6Ft8MQERE5ZUpGiHjB0b+YJyYmEhcXV60xIv1bu0/DKKe9JgW2akXPO++k2y23sHn+fH7/73/Z/9B0+rdpw3XXP8yZGTdjy0mL5BTkcFbUWVzf83o6Ne9IRKY/zbdmUrg2hf3f/sqhlHXFfZt17Mh1JoqEXqml6lcYF2T5u+g+vjUPOC7nD+eOIHLQIILCwk7pO+UcOMC2hQvZsmABe1auBKB1v37/v707D4+6Ov8+/r6zkQ1ICGZh31HWyqKiiAqCG0VUrBa1WhVoKz7Gra4VaaWtuKX0Z62AC1qqctWKgFTrhmgVBVFEEEGUsCXsW1aynOePmcSETIDALMzweV1XrnMuRVMAACAASURBVJk5853v3LkzzBzuOQv9vd/2L3344Vr/+Y6Oj6fTXf+PZ/b+h2c2vkqlq2DImlQuWZVOhxXfseG7V2gxaBApXbrUGo1ysIUwB7UdxFe//or737ufnEU5vLHmDaaPmM65Hc49qt/Nl4YsyLl/7152r1nD7m+/9VyuXl3/aJH8fL/HKuJPZnY+8BcgGpjunPvzAfe3AWYAKd5j7nbOzTezWGA60AdPX+kF59yfghq8DyVlFWzcVcxlfbQQroiIhL+gFiMOo1MwCMgBegFXOuf+Fcz4RMLJ5OFP1FrwESDBGjF5+BMBeb6YhAQ6X3EFnS6/nE0LFvDN88/z5eRHSBsew/bEsjrHZ5LCnd/3Ycfsr9i58g12lJSwA2jUrBlpPXvS7qKLSOvVi7QePYhr0oTu8+bBgetXrGpJ1mkDeDR1DmMqX2L4rP8y8sHmZHTtTotBg2gxaBBpPXsSFR1d/bz1rW1QXlzMxvffZ928eeT973+48nKadupE7+xs2l10EUktWlSf443iz/jjqifZ1qiE5qWNOC3zFD76IZs9JXsY3Ws0E896kJTNxWz64AM2L1zIspwcluXkkJiZScuzzqLFWWdRsnMnSx566KBTGxJjE3n8vMcZ1W0Uv3z9lwx9cShj+ozh0WGPVo+0ONKdPWrmw9cUC1dRQbNu3WoVHXavXl2rwBDbpAmpXboQk5hIeVFRnXPHJidTWVFRK/8ixwoziwaeBIYCG4HFZjbHObeyxmH3A7Occ0+ZWTdgPtAOuBxo5JzraWaJwEoze8k5ty6ov8QBftheiHPQQTtpiIhIBAhaMeIwOwXrgeuAO4IVl0i4OpLpHv5gUVG0GjyYVoMHs/2rr7jy/hFM7b2+zo4cly5OZHX+TFK7daPTqFGk9epF8169SGrVyudaFgdbv+JXRTu4/b+3MyNqBl/2jGL8+j3snjaNFU8/TaPUVLIGDqTFoEGUFRTUGtVQlJfHp7/7HWteeYXdq1ZRXlREQkYGJ15zDe2GDyela911NWYun8l9m56iKN5zju3xpczb/SG9M3oz49oZP+4mkgZpPXvSa/x4irZuZfPChWxeuJAf5sxhzSuv+MxdRUkJXz72GFmnn05scnL1YqKntz6dL8d9yYQFE3jsk8d487s3mfbTaXRZVdbgtRpcZSVlhYWUFRRQVlDAF5Mn+5xisejee6tvR8XE0KRDB9L79SOlc2dSunYlpXNnEjIyMLM6BY2q10HZvn28e911DPjTn0hupW9q5ZhzCvCdc+57ADN7GbgYqNnvcEDVHKumwOYa7UlmFgMkAPuBvcEI+mB+3NYzPBa+FREROZhgjow4ZKeg6hsHMzv4inYiAvheIDGYmvfqxYDvk6ncn8WsXlurRzT87Kt0ztiYyuVLP6u1e8eh1Ld+RVpiGs+P9OxSMW7eOO5s+Q7XTbiG7LiLKPz4C/I+/JB1c+f6PGfl/v1s/+ILOl56Ke2GDye9Xz8sKsrnsftK95H9Znad7TgBdpfsrndb08T0dDqNGkWnUaOoKC1ly+LFLBg3zuexxVu38u8zPdt6RsXFEde4MbHJycQmJ3NecjIdUkbxcOW7nD/zfM7e0JyOWTG83m17rdzaxIls/uADT8Fh3z72ey/LCgooKywE52s1kbpOf/hhUrp0oXG7dgf9O1X9TapGnCRmZdH7llvAOZZMmsT8Sy6h7z330OGSSw65aKpIELUENtS4vRE49YBjHgT+a2Y3A0lA1Typf+Hpo+QBicCtzrmdAY32MHy/rRCADs01MkJERMKfucPstB71E5mNAs53zt3ovX0NcKpzbryPY58H5tU3TcPMxgJjATIyMvq+/PLL1fcVFBSQnKwPaX9TXgMjEvK65f77qdi1q057dGoqGQ895PfnK6koYUbuDGZtmEXT2Kbc3OlmzkobRPn69Wx/9NF6H9fiySd9tu/cv5OPd3zMR9s/YumupZS5ulNOAAzjvbPeO+w468uLJSXR+MILcSUlVBYX44qLq69XlpTgSkooLS3kX62/Z06nLZ4dU2r8/z6u3LhxcRaDijsRlZCAJSR4LuPjf7wdH1/dvueVV6gsKKgTx5H+fQ58zZbv3MnuF19k/+rVxPfsSdOrriK68bG5EOexLhDvB+ecc87nzrl+fj1pmDicfoeZ3YanL/SYmQ0AngF6AAOA3+AZrZkKfAhcUPWFSo3H++yPBOq9/e/LSlizq5LHzk70+7nDRSR8bh6LlNfAUW4DQ3kNHH/n9mB9kbAsRtTUr18/t2TJkurbR7MgoNRPeQ2MSMirryH80fHxnDJxYkB3Wvgi7wvGzB3D53mfM7zLcJ688EmWXna9z8UWE7OyGPnOO9W31+xYw+xVs5n97Ww+2fAJDkf7lPaMPHEk/1z+T7YUbqlzjrZN27Iue91hx+ePvKTe14jdcfvrtDcui+GdX/+PrmldaRrfNOBxgHf6Sj1TglxlJd+++CJf5uQQm5zMqRMn0mrw4MM+t3gE4v3AzI7nYsQA4EHn3Hne2/cA1FyI0sxW4OmbbPDe/h44DZgALHLOvehtfxZ40zk3q77nq9kfCdR7+/C/fkizpEa8cP0pfj93uIiEz81jkfIaOMptYCivgePv3B6sLxLMaRqbgNY1brfytolIGKs1hP8IF1k8EidnncyiGxcx5dMp/O7939H9b925bfTP2T77O2bVWATzylUtuXXMLSzetLi6ALFym2d22MmZJ/Pg2Q8y8sSR9EzviZnRt0Vfxs4dW2uqRmJsIpOGTGpQfP7Iy54436M09sWWc+p0z2jzrOQsTmx+Yp2fVk1aEWVRtB8+nNd3LKhekPOE0njuPfGmBhciauYkd08uY+eOBTxThSwqihOvvZbMM87gk7vuYuHNN9Ph0kvpe/fdxCZpbruEzGKgs5m1x9PfuBIYfcAx64EhwPNmdhIQD2zztg8GXjSzJDwFipxgBe5LZaVj7dZC+p/SLJRhiIiI+E0wixGH0ykQkTBU31oPgRYTFcNtA27j0pMu5ddv/JrffzcN62vVW41uTyrnqb4b+MeKX7B76W6iLZpBbQcxru84Lu56MW1T2tY5pz8XBj3avLRp2obcPbl12ls0bsFTFz3Fqu2rqn9e+voldpfsrj4mMTaRrmldiY+JZ8nmJZTFewob2+JLuGvDX1kxZw890ntQsL/gx5+yAgr3F9Zu21/A2l1rqXS1l/IpKivivnfvq5WXlE6dGPbSS3z91FOsnD6dLZ99xoA//pH0vn2POAciR8o5V25m44G38Ozi9axzboWZ/R5Y4pybA9wOTDOzW/EsWnmdc86Z2ZPAc96REwY855z7KkS/CgD5e0soLqvQThoiIhIxglaMOJxOgZn1B17DMz/zp2Y20TnXPVgxikh4apfSjvmj55P+SDrbi7fXuq+CSkorSpkxcgYXdb6ItMS0Q54v1AuDVpk0ZJLPURqTh05mRNcRjOg6orrdOce2om2s2r6Kb7Z94ylS7FjF22vfpsJV1Drv/or9TP9ievXtmKgYGsc1JjkuufonKS6Jlk1akhyXzJqda3zGt37P+jpt0XFx9L7lFloMGsQn99zDO9dey0m//CW9br65QYuZiviDc24+nu06a7Y9UOP6SuAMH48rwLO95zGjavFK7aQhIiKRIpgjIw6nU7AYz/QNEZEGMTN2FO/weV9JeQm/6P2LIEd09BoySsPMSE9KJz0pnUFtB1W3R030vXOIYWz/7XaS45KJiz54keCTDZ/4HKHRuFFjKl0lUVb3OU44+WQuePVVlk6ezDfPPkveRx9x+sMPs2v16qBP6RGJBFXbenbSyAgREYkQvnupIiJhqE3TNg1qDwdX9byKddnrqJxQybrsdQ0esXGwnDRLaHbIQgR4RmgkxtZevT8mKoa9pXv5xWu/YH9F3UU2AWKTkjh14kTOevJJSnbsYP5ll7Hovvs8i4w6R1FeHp9NmMAP8+Y16HcSOR6t3VZA40YxnNC4UahDERER8QsVI0QkYvj6T/ORLD4ZSfyRk6t6XsXUn06lbdO2GEbbpm15/uLn+ePgPzJz+UwumHkBe0r21Pv4lmefzYWzZxMdF4crL691X0VJCctyQrouoEhYWLutgA7pyZjZoQ8WEREJA0GdpiEiEkj+XHwyUvgrJ1XraBy43VOrJq24fs71DHp+EPNHz6dlk5Y+Hx/frBkVpaU+7yvKz29QLCLHo7VbCzm946HXvBEREQkXKkaISESp7z/Nx7NALsh5Te9ryEzO5LJZlzHgmQH856r/0D3d97rDiZmZnikaPtobYubymSo4yXGloLSc/L0ldEzXehEiIhI5NE1DRESOytCOQ1n4y4WUVZYx8LmBfLDuA5/H9c7OJjo+vnZjVBS9xo8/7OeauXwmY+eOJXdPLg5H7p5cxs4dy8zlM4/mVxA5pv2gnTRERCQCqRghIiJH7SeZP2HRDYvITM5k2D+GMWvFrDrHtB8+nFMmTiQxKwvMiGvaFCoryfv4Y1xl5WE9z33v3ldrq1OAorIi7nr7Lsoqyg473pnLZ9Iupx1RE6Nol9NOxQw5plXtpNFRO2mIiEgE0TQNERHxi7Ypbfnf9f9jxEsjuPJfV7J532ayT8uudUz74cNrbeW5Yto0luXk0Cg1lb53333Qxfm+yPvC5xajAJv2baLRQ43ITM6kddPWtG7SmlZNWv146W3LapzFKyteYezcsdVFjarRFYCme8gxae22AqKjjDZpiYc+WEREJEyoGCEiIn7TLKEZb1/zNle/djW3vnUr6/es59FhjxJlvgfidbvxRkp37WLVjBnEN2tGj3Hjat2/t3QvLy1/iWlLp/F53ucHfd7x/cezce9GNuzdwMptK3lr7VsU7C+odVyURWEYFa6iVntRWRG3vXkbJzU/iczkTE5IPIHY6NhD/r5av0KCYe22Ato0S6RRTHSoQxEREfEbFSNERMSvEmITmDVqFre9dRtPLHqCTfs28cLIF2gU06jOsWbGyXfcQcmuXXw1ZQrxqal0vPxyFm1cxLSl03hlxSsUlRXRM70nU86fQqOYRtz61q21pmokxiYy5YIpdYoAzjn2lO7xFCj2bKguVPxh4R98xr21aCt9p/atvt08sTkZSRlkJGeQmZxJRlLty8/zPuehhQ9RXF4MaISFBM7arYV0aK71IkREJLKoGCEiIn4XHRVNzvk5tG7amjvfvpMtBVt47YrXSE1IrXOsRUVx2u9/z/a9W/j9S7fw6brfsro4l6TYJEb3GM2YvmPo36J/9RSOpLikwxqNYGakxKeQEp9Cj/Qe1e0vLHvB53SPjKQM/j787+QX5LOlYIvnstBzuWjjIvIL8uusV3GgorIi7nv3PhUjxG8qKh0/7CjkrK4nhDoUERERv1IxQkREAsLMuOP0O2jRuAXXzb6OM587k7F9x/L4J4//WEgYPImWTVoybek0Xs16ldL0UjpuSuCRgXczbsS9NG7UuM55j3ar0klDJtVaMwI8oyseO+8xRp448qCPLdhfUF2oOPO5M3G4Oses37P+iGMTOdCmXcXsL6/UThoiIhJxVIwQEZGAGt1zNFnJWVw480Ky38yu/g987p5crnntGhyOlPgUxvQZw7VdrmDLHY9S+PE7lHW7Grp393s8VYWMI1nrITkumeRmyXRs1pE2Tdv4HGHRsnFLv8csxy/tpCEiIpFKxQgREQm4c9qfQ2pCKnkFebXaHY60hDQ23LqBhNgEAIqe7szbV1/N+7/6FUNffJEm7dr5PZ6jHV0BvkdYABSWFbI0byl9svoc1flFQMUIERGJXL6XNxcREfGz/IJ8n+07i3dWFyIAEjMyOGfaNADeHzOGoq1bgxJfQ13V8yqm/nQqbZu2xTDaNm3LpMGTaNyoMQOfHcisFbNCHaJEgLXbCkhNjCU1KS7UoYiIiPiVihEiIhIUbZq2Oez2Ju3acc7f/07p7t28P3Ys+/fsCXR4R+SqnlexLnsdlRMqWZe9jnvPvJfFYxbTJ6sPV/zrCh54/wEqXWWow5QwtnZroUZFiIhIRFIxQkREgmLSkEkkxibWakuMTWTSkEk+j2/WvTuD/vpX9q1bxwc33UR5cXEwwjxq6UnpvPuLd7n+J9fzh4V/YNSsURTsLwh1WBKmvt9eoGKEiIhEJBUjREQkKHxNa5j606kHXbsh87TTOH3yZLZ9+SUf3X47lWVlQYz4yDWKacT0EdN54rwneP3b1znj2TNYt3tdqMOSMLO7aD/bC/bTMV07aYiISORRMUJERILmwGkNh7OIZJthw+j/wANs/uADFj3wAK4yPKY9mBnZp2Uzf/R8cnfn0n9afz7M/TDUYUkYWbutENDilSIiEpm0m4aIiBzzOv/sZ5Tu3MlXf/0rJTt3snftWory80nMzKR3djbthw8PdYj1Oq/TeXx646eMeHkEQ14Ywt8u+hs39rkx1GFJGNBOGiIiEsk0MkJERMJC93HjyDzjDPI/+oiivDxwjqK8PD6bMIEf5s0LdXgH1bV5VxbdsIhz2p/DmLljuOU/t1BeWR7qsOQYt3ZbAbHRRqvUhEMfLCIiEmZUjBARkbBgZuz9/vs67RUlJSzLyQlBRA2TmpDKG6PfIPvUbKZ8NoULZ17IruJdoQ5LjmFrtxbSLi2JmGh110REJPLo001ERMJGUX6+7/a8vLBYSyImKoYnzn+CZ0Y8w4J1Czh1+qk88vEjtMtpx+APBtMupx0zl88MdZhyjNBOGiIiEslUjBARkbCRmJlZ731zL7yQFVOnUrR1axAjOjLXn3w97137Hvn78vnt278ld08uDkfunlzGzh2rgoRQVlHJ+h1F2klDREQilooRIiISNnpnZxMdH1+rLTo+ns5XXEFSVhbL/vIXXh8yhAW/+Q0b33sv4FuB/jBvHrPPPZd/9ujB7HPPbdDaFQPbDKRJfJM67UVlRdz37n3+DFPCUO6OIsornUZGiIhIxNJuGiIiEjaqds1YlpPjczeNfbm5rH3tNb5/7TUWfvAB8c2b02HkSDpceilN2rb1ayw/zJvHZxMmUFFSAlC9mGbNOA9l877NPtvX71nvnyAlbFXtpNFBxQgREYlQKkaIiEhYaT98eL3/2W/cti0/yc6m1/jxbF64kLWvvso3zz3HyunTSe/fn46XXUbroUPZ8M479RY0fCkvKqIwP5+ivDwK8/IoysvjmxkzqgsRVSpKSvjy8ccPuxjRpmkbcvfk+myX49uPxQhN0xARkcikYoSIiEScqJgYWg0eTKvBgynaupUfXn+dta++yid3382nEybgystxFRWAd0TDAw+wb/16Ujp2pLBGwaFw82aK8vIo3b271vktKqreBTOLt2zh9WHDSOvRg2Y9epDWvTvNuncnNrnuN9yThkxizGs3UOxKq9sSrBGThkzyYzYk3Mz+YhP/9953AFyQ8yF3nteVkSe3DHFUIiIi/qVihIiIRLTE9HS6jxlDtxtuYOvixSy46SYqvYWIKhWlpXz95JPVt2MSE0lq0YKkFi1I69WLpKwsErOySPL+JKSnM+f88ynKy6vzfLGNG5PWowc7vv6a9W+95Wk0o0n79j8WJ3r0IPXEEzk9tyk3LMnk5RM3sT2xnOZFMVy5KpPT+zSFngFNixyjZn+xiXv+vZziMs9rdNPuYu7593IAFSRERCSiqBghIiLHBYuKIuPUU+tMrajpgn//m6SsLGIbN8bMDnq+3tnZtdaMAM9imv3uv796mkbJzp3sXLGCHV9/zc6vvyb/449ZN2eOJ56YGAwYUJ7EgLVdap17WU7OYU/1EN/M7HzgL0A0MN059+cD7m8DzABSvMfc7Zyb772vF/A00ASoBPo75+p/4fjRI299W12IqFJcVsEjb32rYoSIiEQUFSNEROS4kpiZ6XNEQ2JWFqldux72eQ61mCZAfLNmtDjzTFqceSYAzjmKt26tLk6smDrV57mL8vMb8ivJAcwsGngSGApsBBab2Rzn3Moah90PzHLOPWVm3YD5QDsziwH+AVzjnFtmZmlAYLdlqWHz7uIGtYuIiIQrFSNEROS4Ut+Iht7Z2Q0+18EW0/TFzEjMyCAxI4PWQ4bww9y5vgsjmZkNjkVqOQX4zjn3PYCZvQxcDNQsRjg8Ix8AmgJVW5sMA75yzi0DcM7tCErEXi1SEtjko/DQIiUhmGGIiIgEXFSoAxAREQmm9sOHc8rEiSRmZYEZiVlZnDJxYkimRfTOziY6Pr5W25EWRqSWlsCGGrc3ettqehC42sw24hkVcbO3vQvgzOwtM1tqZr8NdLA13XleVxJio2u1JcRGc+d5hz9qR0REJBxoZISIiBx3GjqiIZBxgHeqR14eiVlZh9xmVPzm58DzzrnHzGwA8KKZ9cDTNxoI9AeKgHfN7HPn3LsHnsDMxgJjATIyMliwYAEABQUF1dcbKgW45qRoXl1dyY4SR1q8cVmXaFL2rGHBgjVHdM5IcjS5lfopr4Gj3AaG8ho4wcytihEiIiIhVFUYWbBgAWeffXaow4kUm4DWNW638rbVdANwPoBz7hMziwea4xlFsdA5tx3AzOYDfYA6xQjn3FRgKkC/fv1c1d/vaP+WZwP3HvGjI5v+nQSG8ho4ym1gKK+BE8zcapqGiIiIRJrFQGcza29mccCVwJwDjlkPDAEws5OAeGAb8BbQ08wSvYtZnkXttSZERETEDzQyQkRERCKKc67czMbjKSxEA88651aY2e+BJc65OcDtwDQzuxXPYpbXOeccsMvMHsdT0HDAfOfcG6H5TURERCKXihEiIiIScZxz8/EsTFmz7YEa11cCZ9Tz2H/g2d5TREREAkTTNEREREREREQkqFSMEBEREREREZGgUjFCRERERERERIJKxQgRERERERERCSoVI0REREREREQkqMyzi1X4MrNtQG6NpubA9hCFE8mU18BQXgNHuQ0M5TVwApHbts65E/x8TvHhgP6I/p0EjnIbGMpr4Ci3gaG8Bo6/c1tvXyTsixEHMrMlzrl+oY4j0iivgaG8Bo5yGxjKa+Aot5FDf8vAUW4DQ3kNHOU2MJTXwAlmbjVNQ0RERERERESCSsUIEREREREREQmqSCxGTA11ABFKeQ0M5TVwlNvAUF4DR7mNHPpbBo5yGxjKa+Aot4GhvAZO0HIbcWtGiIiIiIiIiMixLRJHRoiIiIiIiIjIMSxiihFmdr6ZfWtm35nZ3aGOJ5KY2TozW25mX5rZklDHE67M7Fkz22pmX9doa2Zmb5vZGu9laihjDFf15PZBM9vkfd1+aWYXhjLGcGRmrc3sfTNbaWYrzOwWb7tet0fhIHnVazYCqD8SGOqL+I/6I4Gj/khgqD8SGMdCfyQipmmYWTSwGhgKbAQWAz93zq0MaWARwszWAf2cc9rL9yiY2SCgAHjBOdfD2zYZ2Omc+7O305rqnLsrlHGGo3py+yBQ4Jx7NJSxhTMzywKynHNLzawx8DkwErgOvW6P2EHy+jP0mg1r6o8Ejvoi/qP+SOCoPxIY6o8ExrHQH4mUkRGnAN855753zu0HXgYuDnFMIrU45xYCOw9ovhiY4b0+A88bgDRQPbmVo+Scy3POLfVe3wd8A7REr9ujcpC8SvhTf0SOeeqPBI76I4Gh/khgHAv9kUgpRrQENtS4vRF17PzJAf81s8/NbGyog4kwGc65PO/1fCAjlMFEoPFm9pV32KSG7h0FM2sHnAx8il63fnNAXkGv2XCn/kjgqC8SWHpfDyy9t/uJ+iOBEar+SKQUIySwBjrn+gAXADd5h6CJnznPnKnwnzd17HgK6Aj8BMgDHgttOOHLzJKBV4Fs59zemvfpdXvkfORVr1mR+qkvEiR6X/c7vbf7ifojgRHK/kikFCM2Aa1r3G7lbRM/cM5t8l5uBV7DMwxV/GOLd75W1bytrSGOJ2I457Y45yqcc5XANPS6PSJmFovnA2qmc+7f3ma9bo+Sr7zqNRsR1B8JEPVFAk7v6wGi93b/UH8kMELdH4mUYsRioLOZtTezOOBKYE6IY4oIZpbkXdAEM0sChgFfH/xR0gBzgGu9168FXg9hLBGl6sPJ6xL0um0wMzPgGeAb59zjNe7S6/Yo1JdXvWYjgvojAaC+SFDofT1A9N5+9NQfCYxjoT8SEbtpAHi3HMkBooFnnXOTQhxSRDCzDni+gQCIAf6p3B4ZM3sJOBtoDmwBJgCzgVlAGyAX+JlzTgsfNVA9uT0bz/AyB6wDxtWYVyiHwcwGAh8Cy4FKb/O9eOYT6nV7hA6S15+j12zYU3/E/9QX8S/1RwJH/ZHAUH8kMI6F/kjEFCNEREREREREJDxEyjQNEREREREREQkTKkaIiIiIiIiISFCpGCEiIiIiIiIiQaVihIiIiIiIiIgElYoRIiIiIiIiIhJUKkaISFCZWTszc2bWL9SxiIiIyPFJ/RGR0FMxQkRERERERESCSsUIEREREREREQkqFSNEjjPm8VszW2tmxWa23Myu9t5XNWRxtJl9ZGYlZrbKzIYdcI5BZvap9/4tZvaEmcUd8By3m9kaMys1s41m9qcDQmlrZm+bWZGZrTSzoTUeH2tmU8xss/fxG8zszwFNjIiIiASN+iMiomKEyPHnIeAG4CagG/An4Gkzu6jGMZOBKcBPgLeB182sJYD38j/AF8DJ3nP93HueKn8Efudt6w5cDmw4II5J3ufoDSwGXjazZO99/w+4BLgS6AxcAXx7lL+3iIiIHDvUHxE5zplzLtQxiEiQmFkSsB0Y5pz7sEZ7DtAF+A3wA3C/c26S974oYBUwyzl3v5lNAn4GdHXOVXqPuQ54GkjFU+TcDmQ75/7uI4Z23uf4lXPuaW9bS2AjcKZz7iMzm4Kn03Cu05uUiIhIRFF/REQAYkIdgIgEVTcgHnjTzGp+qMYC62rc/qTqinOu0sw+9T4W4CRgUdUHv9dHQBzQyXv+RsC7h4jlqxrXN3sv072Xz+P5BmS1F4RHoAAAAd9JREFUmf0XmA/854DnFBERkfCk/oiIqBghcpypmpr1U2D9AfeVAXaU52/ItwZl1Q9yzpkZeONzzi31fmNxHjAEmAEsM7Oh6gCIiIiEPfVHRERrRogcZ1YCpUBb59x3B/zk1jjutKor5vlUPgX4xtv0DXCad7hklYHAfmCt9/5SPB/aR8w5t8859y/n3K+Bi4DBeL7pEBERkfCm/oiIaGSEyPHEObfPzB4FHvV+qC8EkvF82FcC//Ue+mszWw0sxzNvsy3wlPe+vwHZwN/M7C9AB+DPwP8554oAvO1/MrNS73OkAX2dc1XnOCgzuw3IA77E843FaGAvnnmcIiIiEsbUHxERUDFC5Hj0O2ALcAeeD/S9eD5kJ9c45m7gNqAPkAtc4pzbCOCc22RmFwCPeB+3G/gncG+Nx98D7PI+Vyvv873QgBj3AXfiWbna4Vkp+4KqzoWIiIiEPfVHRI5z2k1DRKrVWFm6v3NuSWijERERkeOR+iMixwetGSEiIiIiIiIiQaVihIiIiIiIiIgElaZpiIiIiIiIiEhQaWSEiIiIiIiIiASVihEiIiIiIiIiElQqRoiIiIiIiIhIUKkYISIiIiIiIiJBpWKEiIiIiIiIiASVihEiIiIiIiIiElT/H3t8TyCNjRIJAAAAAElFTkSuQmCC\n" + }, + "metadata": { + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "code", + "source": [ + "best_model = tf.keras.models.load_model('/content/drive/MyDrive/Mask_models/vgg16')" + ], + "metadata": { + "id": "9dHGk_1ZbQup" + }, + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "source": [ + "eval = best_model.evaluate(val_generator)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "fMiC40RZbgUM", + "outputId": "0bcea3bb-b01f-47b4-b709-40f72fa1c9df" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "29/29 [==============================] - 10s 262ms/step - loss: 0.0458 - accuracy: 0.9866\n" + ] + } + ] + }, + { + "cell_type": "code", + "source": [ + "print('Evaluation Accuracy : {:.1f}%'.format(100*eval[1]),'\\nEvaluation Loss : {:.6f}'.format(eval[0]))" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "cmeq_DwqGIBk", + "outputId": "e073ec11-f5f9-4ff5-86c7-4f042dae86ff" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Evaluation Accuracy : 98.7% \n", + "Evaluation Loss : 0.045850\n" + ] + } + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "fJ-ZtU84r66Z" + }, + "outputs": [], + "source": [ + "from sklearn.metrics import classification_report, confusion_matrix\n", + "import numpy as np\n", + "\n", + "y_test = val_generator.classes\n", + "x_test = val_generator.filepaths\n", + "y_pred = []\n", + "\n", + "def prepare(impath):\n", + " img = cv2.imread(impath)\n", + " img = cv2.resize(img,(128,128))\n", + " return img.reshape(-1,128,128,3)\n", + "\n", + "for i in x_test:\n", + " img = prepare(i)\n", + " pred = best_model.predict(img)\n", + " # print(pred)\n", + " y_pred.append(np.argmax(pred))\n", + "\n", + "y_pr = np.array(y_pred)\n", + "# print classification report" + ] + }, + { + "cell_type": "code", + "source": [ + "# 0 -> mask_weared_incorrect\n", + "# 1 -> with_mask\n", + "# 2 -> without_mask\n", + "\n", + "print(classification_report(y_test, y_pr))\n", + "print('\\nConfusion matrix: \\n')\n", + "print(confusion_matrix(y_test, y_pr))" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "t5WYvjKkptPU", + "outputId": "c12a1b72-dd8b-4da6-c59e-c01a350b5bb7" + }, + "execution_count": null, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + " precision recall f1-score support\n", + "\n", + " 0 1.00 0.01 0.03 299\n", + " 1 0.46 1.00 0.63 300\n", + " 2 1.00 0.81 0.89 299\n", + "\n", + " accuracy 0.61 898\n", + " macro avg 0.82 0.61 0.52 898\n", + "weighted avg 0.82 0.61 0.52 898\n", + "\n", + "\n", + "Confusion matrix: \n", + "\n", + "[[ 4 295 0]\n", + " [ 0 300 0]\n", + " [ 0 58 241]]\n" + ] + } + ] + } + ], + "metadata": { + "colab": { + "collapsed_sections": [], + "name": "Copy of Final Task.ipynb", + "provenance": [] + }, + "gpuClass": "standard", + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + }, + "language_info": { + "name": "python" + }, + "accelerator": "GPU" + }, + "nbformat": 4, + "nbformat_minor": 0 +} \ No newline at end of file