11_image_compression_autoencoder_cifar10.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a7280e46",
   "metadata": {
    "id": "a7280e46"
   },
   "outputs": [],
   "source": [
    "from IPython.display import display\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.compose import make_column_transformer\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "from sklearn.metrics import r2_score\n",
    "\n",
    "from keras.models import Sequential, load_model\n",
    "from keras.layers import Dense \n",
    "from keras.datasets import cifar10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ece7314e",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ece7314e",
    "outputId": "0236719a-74c4-4e9b-8720-d54dced5cdb1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz\n",
      "170498071/170498071 [==============================] - 6s 0us/step\n"
     ]
    }
   ],
   "source": [
    "(X_train, y_train), (X_test, y_test) = cifar10.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "26b3f8b2",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "26b3f8b2",
    "outputId": "09cdf35e-1ef3-4d38-aba4-20c0090226e2"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 32, 32, 3)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "0e8f6b57",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "0e8f6b57",
    "outputId": "fcb736e4-3ab5-4321-d353-6cdf1290d72a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 1)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "418f3d1c",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "418f3d1c",
    "outputId": "95da97e1-d091-459e-f688-510443ba5148"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 32, 32, 3)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "7c065e96",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "7c065e96",
    "outputId": "119c99d3-656d-491f-fe9d-b49f035914e0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 1)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "6dc3af9e",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 448
    },
    "id": "6dc3af9e",
    "outputId": "55c2639b-2019-4a40-a836-722758933c77"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f1534cc5fc0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_train[5])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4981ef7f",
   "metadata": {
    "id": "4981ef7f"
   },
   "source": [
    "### Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "40667455",
   "metadata": {
    "id": "40667455"
   },
   "outputs": [],
   "source": [
    "# Normalize and reeshape images \n",
    "X_train_final = X_train.reshape(-1, 32*32*3) / 255\n",
    "X_test_final = X_test.reshape(-1, 32*32*3) / 255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "35239126",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "35239126",
    "outputId": "a959342e-3973-4a1c-be09-a8d0ef446be3"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 3072)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_final.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9f3662fc",
   "metadata": {
    "id": "9f3662fc"
   },
   "source": [
    "### Modelling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "f3a008c3",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "f3a008c3",
    "outputId": "f6761708-5b8a-448e-9bbb-19a451175f4d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 1024)              3146752   \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 512)               524800    \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 128)               65664     \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 512)               66048     \n",
      "                                                                 \n",
      " dense_4 (Dense)             (None, 1024)              525312    \n",
      "                                                                 \n",
      " dense_5 (Dense)             (None, 3072)              3148800   \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 7,477,376\n",
      "Trainable params: 7,477,376\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "\n",
    "# Encoder layers\n",
    "model.add(Dense(1024, input_dim=3072, activation='relu'))\n",
    "model.add(Dense(512, activation='relu'))\n",
    "\n",
    "# Bottleneck layer\n",
    "model.add(Dense(128, activation='relu'))\n",
    "\n",
    "# Decoder layers\n",
    "model.add(Dense(512, activation='relu'))\n",
    "model.add(Dense(1024, activation='relu'))\n",
    "\n",
    "# output layer\n",
    "model.add(Dense(3072, activation='sigmoid'))\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "ec39d352",
   "metadata": {
    "id": "ec39d352"
   },
   "outputs": [],
   "source": [
    "model.compile(loss=\"BinaryCrossentropy\", optimizer=\"Adam\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "7077bdf5",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "7077bdf5",
    "outputId": "cad087b5-483b-4344-bd58-6f901585c992"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "1563/1563 [==============================] - 15s 6ms/step - loss: 0.6128 - val_loss: 0.6010\n",
      "Epoch 2/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5952 - val_loss: 0.5938\n",
      "Epoch 3/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5910 - val_loss: 0.5915\n",
      "Epoch 4/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5902 - val_loss: 0.5912\n",
      "Epoch 5/50\n",
      "1563/1563 [==============================] - 10s 7ms/step - loss: 0.5898 - val_loss: 0.5906\n",
      "Epoch 6/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5893 - val_loss: 0.5901\n",
      "Epoch 7/50\n",
      "1563/1563 [==============================] - 9s 6ms/step - loss: 0.5886 - val_loss: 0.5905\n",
      "Epoch 8/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5882 - val_loss: 0.5895\n",
      "Epoch 9/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5880 - val_loss: 0.5898\n",
      "Epoch 10/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5878 - val_loss: 0.5891\n",
      "Epoch 11/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5876 - val_loss: 0.5895\n",
      "Epoch 12/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5874 - val_loss: 0.5894\n",
      "Epoch 13/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5873 - val_loss: 0.5902\n",
      "Epoch 14/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5871 - val_loss: 0.5895\n",
      "Epoch 15/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5870 - val_loss: 0.5901\n",
      "Epoch 16/50\n",
      "1563/1563 [==============================] - 9s 5ms/step - loss: 0.5869 - val_loss: 0.5895\n",
      "Epoch 17/50\n",
      "1563/1563 [==============================] - 9s 6ms/step - loss: 0.5868 - val_loss: 0.5893\n",
      "Epoch 18/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5866 - val_loss: 0.5897\n",
      "Epoch 19/50\n",
      "1563/1563 [==============================] - 9s 6ms/step - loss: 0.5866 - val_loss: 0.5897\n",
      "Epoch 20/50\n",
      "1563/1563 [==============================] - 15s 10ms/step - loss: 0.5865 - val_loss: 0.5894\n",
      "Epoch 21/50\n",
      "1563/1563 [==============================] - 11s 7ms/step - loss: 0.5864 - val_loss: 0.5895\n",
      "Epoch 22/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5863 - val_loss: 0.5898\n",
      "Epoch 23/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5862 - val_loss: 0.5897\n",
      "Epoch 24/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5861 - val_loss: 0.5895\n",
      "Epoch 25/50\n",
      "1563/1563 [==============================] - 7s 5ms/step - loss: 0.5860 - val_loss: 0.5896\n",
      "Epoch 26/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5859 - val_loss: 0.5895\n",
      "Epoch 27/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5858 - val_loss: 0.5895\n",
      "Epoch 28/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5857 - val_loss: 0.5902\n",
      "Epoch 29/50\n",
      "1563/1563 [==============================] - 9s 6ms/step - loss: 0.5856 - val_loss: 0.5898\n",
      "Epoch 30/50\n",
      "1563/1563 [==============================] - 7s 5ms/step - loss: 0.5856 - val_loss: 0.5899\n",
      "Epoch 31/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5855 - val_loss: 0.5899\n",
      "Epoch 32/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5854 - val_loss: 0.5898\n",
      "Epoch 33/50\n",
      "1563/1563 [==============================] - 7s 5ms/step - loss: 0.5853 - val_loss: 0.5898\n",
      "Epoch 34/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5852 - val_loss: 0.5903\n",
      "Epoch 35/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5852 - val_loss: 0.5901\n",
      "Epoch 36/50\n",
      "1563/1563 [==============================] - 7s 5ms/step - loss: 0.5851 - val_loss: 0.5901\n",
      "Epoch 37/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5851 - val_loss: 0.5899\n",
      "Epoch 38/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5850 - val_loss: 0.5898\n",
      "Epoch 39/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5849 - val_loss: 0.5899\n",
      "Epoch 40/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5849 - val_loss: 0.5899\n",
      "Epoch 41/50\n",
      "1563/1563 [==============================] - 7s 5ms/step - loss: 0.5848 - val_loss: 0.5902\n",
      "Epoch 42/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5848 - val_loss: 0.5900\n",
      "Epoch 43/50\n",
      "1563/1563 [==============================] - 9s 6ms/step - loss: 0.5847 - val_loss: 0.5902\n",
      "Epoch 44/50\n",
      "1563/1563 [==============================] - 7s 5ms/step - loss: 0.5847 - val_loss: 0.5904\n",
      "Epoch 45/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5846 - val_loss: 0.5907\n",
      "Epoch 46/50\n",
      "1563/1563 [==============================] - 9s 5ms/step - loss: 0.5845 - val_loss: 0.5899\n",
      "Epoch 47/50\n",
      "1563/1563 [==============================] - 7s 5ms/step - loss: 0.5845 - val_loss: 0.5904\n",
      "Epoch 48/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5844 - val_loss: 0.5906\n",
      "Epoch 49/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5844 - val_loss: 0.5909\n",
      "Epoch 50/50\n",
      "1563/1563 [==============================] - 8s 5ms/step - loss: 0.5843 - val_loss: 0.5902\n"
     ]
    }
   ],
   "source": [
    "result = model.fit(X_train_final, X_train_final, verbose=1, epochs=50, validation_data=(X_test_final, X_test_final))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "716ad17a",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 472
    },
    "id": "716ad17a",
    "outputId": "206adabb-281b-44ae-e11a-787b8007f05e"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(result.history[\"loss\"])\n",
    "plt.plot(result.history[\"val_loss\"])\n",
    "plt.title(\"Model Loss\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.legend([\"Train\", \"Test\"], loc=\"upper right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "f6fb9896",
   "metadata": {
    "id": "f6fb9896"
   },
   "outputs": [],
   "source": [
    "model.save(\"models/image_compression_autoencoder_model_cifar10.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "d10653a3",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "d10653a3",
    "outputId": "e7476528-2143-4e9e-fe16-f884402324d2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 1s 3ms/step\n"
     ]
    }
   ],
   "source": [
    "my_model = load_model(\"models/image_compression_autoencoder_model_cifar10.h5\")\n",
    "predict = my_model.predict(X_test_final)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "c5ac3c6f",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 843
    },
    "id": "c5ac3c6f",
    "outputId": "71180ed2-22fd-490d-cdcc-b326117886b0"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaAAAAGdCAYAAABU0qcqAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAwG0lEQVR4nO3dfXDV9Zn//9e5z/0JAZKQEiyI4i3sLKs0Y+taYQX2N/600h1tO7PYdXR0g7PKdtuy02p1dyZd/U5r26H4x+7KdqZo607R0dnqKpb4bRfchcqPWisVFgWEBAjk7iTn/v37w6/5bhT0fUHCO4nPx8yZgeTKlffn5uTKyfmc14k455wAADjHoqEXAAD4eGIAAQCCYAABAIJgAAEAgmAAAQCCYAABAIJgAAEAgmAAAQCCiIdewPuVy2UdPnxYtbW1ikQioZcDADByzmlgYEAtLS2KRk//OGfCDaDDhw+rtbU19DIAAGfp4MGDmj179mk/P24DaP369Xr44YfV1dWlRYsW6Qc/+IGuvPLKj/y62tpaSdL/+smvVFlV4/fNSiXvdZ043u1dK0m5XNa79pNz55l619fVedfGY7ZHg8lEzL/2Q35DOZWEsT4e8U97KpX897ck1VT5n8LWfWipj0X897ck9faeNNXX1HreFyQl4glT77hh7ZGobR8Wy3nvWuNpZRKN2JoPDQ2b6uNx/32YSlWYehfy/vuwWPCvlaQKw1oiUf9tHBgY0OWXXDDy8/x0xmUA/eQnP9HatWv16KOPasmSJXrkkUe0fPly7dmzR42NjR/6te/92a2yqkaV1R+++BGlovfaKoYGvWsl206v8l3ve/U1/vWJmO0OZBpA1t7jOYCKth+eNdX+9YnxHECG80SSioZzVtJH3pH/p0SCAfSB3sYBFIvZfjTG4/71FRW2AZS3DKB8ztS7sqLSu9bys3Dkaz7iaZRxOeTf+c53dPvtt+vLX/6yLrnkEj366KOqqqrSP//zP4/HtwMATEJjPoDy+bx27typZcuW/d9vEo1q2bJl2rZt2wfqc7mc+vv7R90AAFPfmA+g48ePq1QqqampadTHm5qa1NXV9YH6jo4OpdPpkRsXIADAx0Pw1wGtW7dOfX19I7eDBw+GXhIA4BwY84sQZsyYoVgspu7u0VebdXd3q7m5+QP1qVRKqVRqrJcBAJjgxvwRUDKZ1OLFi7Vly5aRj5XLZW3ZskVtbW1j/e0AAJPUuFyGvXbtWq1evVp/9Ed/pCuvvFKPPPKIMpmMvvzlL4/HtwMATELjMoBuvvlmHTt2TPfdd5+6urr0B3/wB3ruuec+cGECAODja9ySENasWaM1a9ac8ddXV6VUVeX33FDU+W9GLmN7vqmcH/KurUjaXqRXXWl4Fb8xFi8q/3SIVNz2l9jKpK0+qrJ3bc6QaiFJqbj/i/qSCeO6Dfvc8kr4d9diq48aXswbMexvSUolk961xtcsKzNU8K61Ph+QNKzbybi/jS+4TRheiGp9oXAh5//i0rjxBbeVluffDdmc+aTfNga/Cg4A8PHEAAIABMEAAgAEwQACAATBAAIABMEAAgAEwQACAATBAAIABMEAAgAEwQACAAQxblE8ZyuuouKRoletJXYmGbPFlCSihkibqN9631NhWEsiZosGyQ37RwjFYrZ4ooq4//vIS1Ihl/Wujcq2D13Rv7eL2E73kvzjb5IJ2z6xROtIkpz/eRgx/l5ZKvvH5QwN+Z9XktRz7Jh3bdOMaabeEUNcTixpO/Yxw7GXpJjheBoToRQ3bGeuZLv/xA0/VwoF/97Ocx08AgIABMEAAgAEwQACAATBAAIABMEAAgAEwQACAATBAAIABMEAAgAEwQACAATBAAIABMEAAgAEMWGz4JLRkpKeOWzlYt67b0z+uVeSlIga8tqMvaMl/1ytZMKW1xaJ+WeHJaL+++/dettpU47494+Wc6bexawhqy9Wbeqdzfuvu6rKlgUXM+R7SZLKhgxDZ8sxy2T98/R27vy1qXfBkEk4re4KU+9Uyv/3Z2OUoiLOlhmpsv95GDXmzEUMOYDlsjFL0bBuZ+jtW8sjIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEBM2iicRjygZ98vPcBH/nI1E1BaDoZJ/NExMthiMiKF3QjFT70LRP16lVLbllMTqkqb6iDNEFJVtsUDloiEypWSLMxrs7/WuramqMPWOGuNyinn/cyWesN2te4f843JO9PvXSlJl3P933Lzt7qN8wf/Yx5O2/e2MUTylkv85XjREh0lS3nDsk3HbsXeGiKdyyRAJ5FnLIyAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEBM2Cy4VKSkV8csTKkX8Q6QSUVvGUyHnn6kWNWbBubKhd8R2qOJR/7XEY8YsuIgh202SM2TeScaMtLL/WkqyrXtwoN+79oDhPJGkqCEjTbJlk7XWVZl69xw75l37/+3ebeq98NJLvWvLxvMwV/LPVKtwCVPvsjGTcHjIvz4Zt/0MKhb88/dicduxLxT9f07kcv7ryOeHvep4BAQACGLMB9C3vvUtRSKRUbeLLrporL8NAGCSG5c/wV166aV68cUX/+83MUaEAwCmvnGZDPF4XM3NzePRGgAwRYzLc0BvvvmmWlpaNG/ePH3pS1/SgQMHTluby+XU398/6gYAmPrGfAAtWbJEGzdu1HPPPacNGzZo//79+sxnPqOBgYFT1nd0dCidTo/cWltbx3pJAIAJaMwH0MqVK/Vnf/ZnWrhwoZYvX65/+7d/U29vr37605+esn7dunXq6+sbuR08eHCslwQAmIDG/eqA+vp6XXjhhdq7d+8pP59KpZRKpcZ7GQCACWbcXwc0ODioffv2adasWeP9rQAAk8iYD6CvfOUr6uzs1FtvvaX/+I//0Oc+9znFYjF94QtfGOtvBQCYxMb8T3CHDh3SF77wBfX09GjmzJn69Kc/re3bt2vmzJmmPrFiTrGiX3xG2RBVES0aIzb6DFflGaIqJMlF/aNhYpW2Q5U0RNQk4zFT70ghY6ovWfZLybiWuH98i4vYjn0m0+dd291tO/bVdTWmehf1/13RGV93lx/0X3tFwvbn8mO9vd61v37NFvNTnfI/V+bPm2fqHTdGQuWGTn2R1alUxm29yzm/WBtJKhVtMT8lS0JR1vCzMDvoVTbmA+iJJ54Y65YAgCmILDgAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBDj/nYMZyoVcaqI+GUmRZx/tpI1Cy7lSt61NWX/XDJJSss/yyraZ8tfS5X9111hi6ZSdMg/m0qSoln/rLFk1PjWHCX/fZ7vtx372mr/tUxraDD13n+oy1T/3wf963+/d4up98njvd61g1ljlmLht961Mdl6FwxZfZctuNDU+//9f1aY6j/RNN27Nlfhf9+UpGzG/76fz9jOqzrnn9EZGfbPu4t4rplHQACAIBhAAIAgGEAAgCAYQACAIBhAAIAgGEAAgCAYQACAIBhAAIAgGEAAgCAYQACAICZsFM87Bw6oqqrKq7ZQ8I/wGOj3j5OQpFIh5137zjvvmHqfTCW8azOD/abejdP9o2FqqitMvWNxW5RIvlD0ro0nK029o/Gkd23GEAkkSdmoIVrJ2e5KBw4fN9XvP3TCuzaT998nklSRbvSujVSXTb1rDLXVSdvvw0fe/r137eHD3abe//t//8pUf/EF87xrZ9bXmXoPD/Z612b6e0y9Cxcv8K4d7Dvpvw7PuC4eAQEAgmAAAQCCYAABAIJgAAEAgmAAAQCCYAABAIJgAAEAgmAAAQCCYAABAIJgAAEAgmAAAQCCmLBZcP/xyn8plUp51UYiMe++5bItx2x4OONd+1bXYVNvS9RY3PirwrS0f95UdYUtOyxlWLckJeL+xyfueczfE43759gNZf0zAyUpbtiHLmZbd9eJQVN9oex/AlTV1pt6S/5ZfflBW55eVP4nSzbrf1+TpLpa/+PzqcWXm3pn+vyz9yQpm8161x444J+pJkn79u3zrh0uOlPvt3v8MtskaXjI//jkcn4ZmjwCAgAEwQACAATBAAIABMEAAgAEwQACAATBAAIABMEAAgAEwQACAATBAAIABMEAAgAEwQACAAQxYbPgfrPvLcUTCa/aqspa777O+edeSVKu6J99lZ423dQ7lfTPMcsbc7KODfpnU8UitnC32opqU32xVPCujSRsvxPFYv77MBK3rTuV8Tv/JClf6Df1PnHCljUm+Wd8GQ+n8iW/3C5JGsj4Z4dJUn7Yv3frzAZT7+nTmr1rM5k+U+8TJ4/Z1lLvfx7+0aJLTb0PHXnHu7Zv2D93UZLeONTjXRuN+vfO5/3u8zwCAgAEYR5AL7/8sq6//nq1tLQoEonoqaeeGvV555zuu+8+zZo1S5WVlVq2bJnefPPNsVovAGCKMA+gTCajRYsWaf369af8/EMPPaTvf//7evTRR/XKK6+ourpay5cvN8WVAwCmPvNzQCtXrtTKlStP+TnnnB555BF94xvf0A033CBJ+tGPfqSmpiY99dRTuuWWW85utQCAKWNMnwPav3+/urq6tGzZspGPpdNpLVmyRNu2bTvl1+RyOfX394+6AQCmvjEdQF1dXZKkpqamUR9vamoa+dz7dXR0KJ1Oj9xaW1vHckkAgAkq+FVw69atU19f38jt4MGDoZcEADgHxnQANTe/e11+d3f3qI93d3ePfO79UqmU6urqRt0AAFPfmA6guXPnqrm5WVu2bBn5WH9/v1555RW1tbWN5bcCAExy5qvgBgcHtXfv3pH/79+/X7t27VJDQ4PmzJmje+65R3//93+vCy64QHPnztU3v/lNtbS06MYbbxzLdQMAJjnzANqxY4c++9nPjvx/7dq1kqTVq1dr48aN+upXv6pMJqM77rhDvb29+vSnP63nnntOFRX+URWSNFh0ikX84kdc2f+BXFVVjWkdlYaol9mt55t6FzzjKiTp2Gku4jid4z3+ERtNTY2m3qkZs031mV7/tZSjZVPv9LSmjy76P1KpaabeWf/Do6Gi7erNimrbn5pLhUHv2likZOqdjKW8axNJW9RLocK//so/tEXUXHhei3dtNm+Lstq/z/bzat+e171r26643NS7tdV/Ow/sftvUu1Dyj3gql/xjzAoFv3PQPICuueYaOXf6RUciET344IN68MEHra0BAB8jwa+CAwB8PDGAAABBMIAAAEEwgAAAQTCAAABBMIAAAEEwgAAAQTCAAABBMIAAAEEwgAAAQZijeM6VeKpa8UTSq3Zmo39WUkXSNnOPHz/kXZvJDJh6qxzxLs0W/HOYJCk989Rvf3Eqn5g739S7Nm3LVKub4Z8113PipKl3qex/CnvGU40YHvbPDxsa8s9qk6R8Ydi2GPkH0yWTtrt1Rarauzbh8qbejYa3V5k5zZaPV5Hwvy/PNGQGSlJdMmGq7zlwwLv27X1vmXo3N8zwru3r3m7qnWiY6V2bjxnua1G/n208AgIABMEAAgAEwQACAATBAAIABMEAAgAEwQACAATBAAIABMEAAgAEwQACAATBAAIABDFho3jS6elKJFNetTFDREQulzWtI2KY0Sd6ek29+/v941tiCb99MVJfjnnXvv1Ot6l3Xb8tRiadrveujcUqTL1zWf9omEjEFmeUShjuHtVVpt6VznY8o3H/2Ca5sql3daX/2hPOPxJIkmZP94/5qUr6n7OSlOnv9a4tGqOSIs5UrrmGOKvfvfHfpt4XXrjAv7hkO8ePHH7HuzY1rcG7tljwu1/yCAgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQxITNgoslUt75Z0PD/nlgMWPIUyzun01WKtnmeTxe411bdrbeyVStd+2MGbNMvWtqKk31FZX++zCdsmXBxRNJ71oXMeSpSXIl/3OlWLRlpKXr/I+9JEWj/mspl/zvD5IUd/715ZwtUy2d8t/nrpgz9S6V/OvzRVvO3LAhY1CSqmrT3rVvd/WYer++79+9a3M5W05jIeefHedi/vuwWPC7P/AICAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQxISN4mmY0ahkyi/ypVwoe/etqUyY1lEu+UdbJKK2GJnGxhbv2kjctu5khX9cTtIYf1NRYTttYnH/33OscTmRmKHe2DsW8V/3UMYWURN1/uesJKUS/vvcGWJ7JGmozz8a5p233jT1PpHw3+f1lbbzqml6vXdtRUWVqXc27x9RI0ku7hcbJknxqjpT72OHDnvXts6aaepdm/c/D/sNsT1Fz7sOj4AAAEEwgAAAQZgH0Msvv6zrr79eLS0tikQieuqpp0Z9/tZbb1UkEhl1W7FixVitFwAwRZgHUCaT0aJFi7R+/frT1qxYsUJHjhwZuT3++ONntUgAwNRjvghh5cqVWrly5YfWpFIpNTc3n/GiAABT37g8B7R161Y1NjZqwYIFuuuuu9TTc/qrbHK5nPr7+0fdAABT35gPoBUrVuhHP/qRtmzZon/4h39QZ2enVq5cqVKpdMr6jo4OpdPpkVtra+tYLwkAMAGN+euAbrnllpF/X3755Vq4cKHOP/98bd26VUuXLv1A/bp167R27dqR//f39zOEAOBjYNwvw543b55mzJihvXv3nvLzqVRKdXV1o24AgKlv3AfQoUOH1NPTo1mzZo33twIATCLmP8ENDg6OejSzf/9+7dq1Sw0NDWpoaNADDzygVatWqbm5Wfv27dNXv/pVzZ8/X8uXLx/ThQMAJjfzANqxY4c++9nPjvz/vedvVq9erQ0bNmj37t36l3/5F/X29qqlpUXXXXed/u7v/k6plH9WkiRVVdYq6ZnfVMjmvftWVttyz+rrGr1ry0VbBlc8mfSuraypNfV2kZh3bTRmOw3Kzr+3JEUtD7SNj8mdod7Jlr9WLPrnABZLQ6be/T3HTfWWI5QwZsEN9h3zrj1y2D+XTJKaGvz/pF5fPcPUe8iQY1Y25BFKUtH4o9GV/Pf5J2bbnuNecME879o/uMS/VpJ+/98HvWtf/c3vvGvz+ZxXnXkAXXPNNXLu9Dv7+eeft7YEAHwMkQUHAAiCAQQACIIBBAAIggEEAAiCAQQACIIBBAAIggEEAAiCAQQACIIBBAAIggEEAAhizN8PaKxksjkVPIO+aiv9c9Jixtyzo8dO/26u79ff12vqXS77z//5Fy4w9a5v8M/ViiVs2W4R2eqLJf/MLt8MqfcM5TPetdmcLa+tmPd/d95IqWDq7XK27axOJrxr6+sbTL0rkzO9a+MRW85cfY1fnqMkpWv9ayUpb9iHQ4b72ru9bcczGil6105L295ypirlv/ZDB9829Y4ZDuelCy7wrh0e9stR5BEQACAIBhAAIAgGEAAgCAYQACAIBhAAIAgGEAAgCAYQACAIBhAAIAgGEAAgCAYQACCICRvFk4onlEz4xY/0HD/q3XffyeOmdZRK/nEf9dOmmXrPmtXkXZsv2qJBCvmsd23ZlUy9+4f8428kaXjYPwKnVLRF1MSi/lkiyYTt9y1L/E1FdaWpd2XCdtfLDg1615blH30kSdU1Nd61sUjE1DsZ849tisVsxydhOD7Zon9UjiRFDOuWpIhhnxcKeVPvQz0nvWuHMn2m3vF4yru2edZs79pIzG9/8wgIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEMSEzYLr6z2hRKrCq/bIO4e9+1ZVV5nWcdEll3vXNsxoNPWuqvLPD8sO2/LXTp484V1bKNjy14acLcuqqsrvOEpSus4/m0qSqlP+9ZWG7DBJihtyz0olW1ZfsWjbh4WCf15fNmrMPZP/dkajtoy0Usk/q6/gXypJiseS3rWu7J+NKEnZnK2+55h/xuTxHlse5cDAgHftyd5eU+/qqmrv2lTtdO/abNZv//EICAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQxISN4pk2Y6aSFX6xOdMMETjxmC1KJF7hHyMzMDho6j042O9dm0rZYmQKBf+ol3LRFiPT0jTTVJ+q8I9MiUVteSyu7B87k8kOm3pn+/0jUHoN0UeS1HPimKl+2BDFdPHFC0y9E/X13rX+oT3vikX9vyJbtEUI5TL+x+dQ10FT72PHbccnn/e/vw1lbLFafb193rXJmO1HuuVn1paXXvKuLXr+TOEREAAgCNMA6ujo0BVXXKHa2lo1Njbqxhtv1J49e0bVZLNZtbe3a/r06aqpqdGqVavU3d09posGAEx+pgHU2dmp9vZ2bd++XS+88IIKhYKuu+46Zf7HQ8p7771XzzzzjJ588kl1dnbq8OHDuummm8Z84QCAyc30B8Pnnntu1P83btyoxsZG7dy5U1dffbX6+vr0T//0T9q0aZOuvfZaSdJjjz2miy++WNu3b9enPvWpsVs5AGBSO6vngPr63n1yrKGhQZK0c+dOFQoFLVu2bKTmoosu0pw5c7Rt27ZT9sjlcurv7x91AwBMfWc8gMrlsu655x5dddVVuuyyyyRJXV1dSiaTqn/fVTVNTU3q6uo6ZZ+Ojg6l0+mRW2tr65kuCQAwiZzxAGpvb9drr72mJ5544qwWsG7dOvX19Y3cDh60XS4JAJiczuh1QGvWrNGzzz6rl19+WbNnzx75eHNzs/L5vHp7e0c9Curu7lZzc/Mpe6VSKaUMb6sMAJgaTI+AnHNas2aNNm/erJdeeklz584d9fnFixcrkUhoy5YtIx/bs2ePDhw4oLa2trFZMQBgSjA9Ampvb9emTZv09NNPq7a2duR5nXQ6rcrKSqXTad12221au3atGhoaVFdXp7vvvlttbW1cAQcAGMU0gDZs2CBJuuaaa0Z9/LHHHtOtt94qSfrud7+raDSqVatWKZfLafny5frhD384JosFAEwdpgHk3EfndFVUVGj9+vVav379GS9KkgrOKeLx/d79nv7PIcXjtky1kit718YithyzeMz/L6CGSC1JUoUhf204459jJUnDff4ZXJI0bCiPJ23XxUQT/vWuZMsa2/O7171rD7z1lql3sWTb586VvGtbZp36+dbTaUinvWuHh4ZMvS31vSd7Tb17Tvb4ryNvywEsGc+VIcN29hlfahKV/8+Vqrjtaf2uI0f8a09zJfOplMt+5ytZcACAIBhAAIAgGEAAgCAYQACAIBhAAIAgGEAAgCAYQACAIBhAAIAgGEAAgCAYQACAIM7o7RjOhX17f6940i9i55JLL/HuW2mIqJGksn8Sj6Ky5eX4xlVIUvfRo6bemf4+79rcsDGmpGiLKbHEmsyb/0lT75mNM/zXYTmYkhKG2KZ0us7UO2U8D2Mx/9psLmvq/caePd61g5lBU2/LWgrG86rsGdUlSZkBY3yU8T4xNJTxrs3nbTFMKUO8Tv/R46bevb293rWlsv/+LnvW8ggIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEMSEzYIr5Abkyn6ZSdnBXu++0ZIth8nJP/8oGrPtzlKx4F375pu/N/Ue7Ov1rk0mbOtOpCpM9XFDkFm56J+PJ0nRoiHfreR/LCVpekOD/zpsMYAaGrZlqg0b6g8ePGTqbVl7xPgrq4v6f8FQ3pZh12fIMcv0+GcjSlLCkL8mSUXDfblYsp3jmd5+/97D/pl0klQyrcVy/yELDgAwgTGAAABBMIAAAEEwgAAAQTCAAABBMIAAAEEwgAAAQTCAAABBMIAAAEEwgAAAQUzYKJ6KeFTxuN98zBtiSiritsyUSNQ/RiYas83zqCECp66uxtS7IuG/7prqKlPvWEXKVF9V4R/dUyz4R5pI0ptvvOFd23fihKl3X2bAu7bkbPEqiaT/8ZGkuOHcSiWTpt6RqH/EylB22NT72Ike/945WxRPzHDfnFZXb+qdz9rWYolWKhZs50rZFJdjzISK+NdHDDlMkQhRPACACYwBBAAIggEEAAiCAQQACIIBBAAIggEEAAiCAQQACIIBBAAIggEEAAiCAQQACIIBBAAIYsJmwUUjMUU9s55KxbJ330jElsFVLvrnMOVy/tlhklQq+ueeVcZthyqaSHjXDmcypt65E4dN9QeH/HOyysWiqXfE+eeYJQz7RJJicf8Mu0SF7byKGu95+bz/fhk8actry2b9j082O2TqbUkmq4jafh8uZPP+tbIdn2Fj5t3wsH99uez/80qSIlH/vVg05ONJkiv57/Nkwn8dvvl1PAICAARhGkAdHR264oorVFtbq8bGRt14443as2fPqJprrrlGkUhk1O3OO+8c00UDACY/0wDq7OxUe3u7tm/frhdeeEGFQkHXXXedMu/7E87tt9+uI0eOjNweeuihMV00AGDyM/0l+rnnnhv1/40bN6qxsVE7d+7U1VdfPfLxqqoqNTc3j80KAQBT0lk9B9TX1ydJamhoGPXxH//4x5oxY4Yuu+wyrVu3TkNDp3/iMpfLqb+/f9QNADD1nfFVcOVyWffcc4+uuuoqXXbZZSMf/+IXv6jzzjtPLS0t2r17t772ta9pz549+tnPfnbKPh0dHXrggQfOdBkAgEnqjAdQe3u7XnvtNf3yl78c9fE77rhj5N+XX365Zs2apaVLl2rfvn06//zzP9Bn3bp1Wrt27cj/+/v71draeqbLAgBMEmc0gNasWaNnn31WL7/8smbPnv2htUuWLJEk7d2795QDKJVKKZVKnckyAACTmGkAOed09913a/Pmzdq6davmzp37kV+za9cuSdKsWbPOaIEAgKnJNIDa29u1adMmPf3006qtrVVXV5ckKZ1Oq7KyUvv27dOmTZv0p3/6p5o+fbp2796te++9V1dffbUWLlw4LhsAAJicTANow4YNkt59sen/9Nhjj+nWW29VMpnUiy++qEceeUSZTEatra1atWqVvvGNb4zZggEAU4P5T3AfprW1VZ2dnWe1oPcM9vUolkh61Q4P9Hr3PXrYr+d7ctmcd22p6F8rSYWCIcuqYMtI+6hj9T9FDVlTkpRI+OfjSVI87n+1fyxmy7KKJ/zrI7bNVLHkn9WXzfgfS0nK5Wz5ewP9/lljznaqqLrWP/MuZsxrc4acxlzGljNXNGQp9uVsx8eS7SZJpbL/fSJiSsiTys6WHWcRj/vnI0bK/ieW7xaSBQcACIIBBAAIggEEAAiCAQQACIIBBAAIggEEAAiCAQQACIIBBAAIggEEAAiCAQQACOKM3w9ovHUf3KdozG95ruwfVVEq2WJkIoaYmnjKP9ZCkiIx/94RY45M0jPGSHr3LdQtrGspG45PsWjLkRkc9I9jyedtvcvOfzujEdt5VTbE/EhSMuV/jBpbWky9M4N93rX9vSdNvYt5/+10xmNvibQZyltjfsYv+sqYxGPazoQxVism//vm0NCAd63vfZ5HQACAIBhAAIAgGEAAgCAYQACAIBhAAIAgGEAAgCAYQACAIBhAAIAgGEAAgCAYQACAIBhAAIAgJmwWXKycVTQS86otl/zzjMrWvClDtlIpatudUedfb4xfU66U864tFmw5WZaMNMmev2cRj/vvw0TSPx9PkmJx/2y/uCULTFKp6H/OSlJF0n87U5UpU++TPf7nSmZg0NQ7EfW7D0tSLGL7fTifM5zjzra/nWzH05KPGI3atjNiOLcq4v77W5IG+3u9a4cy/pmBznN/8wgIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABDEhI3iKZcKkvziHJwhqsI5WxSPK/tHbLiCMY7FEFFjTOJRxBD3UYrZ4jtiCVukTSrlHw0TM8aURA1rtx0dyZX9j0+p4B8LI0ml4WFTfT7hvw+HhzOm3plB/3gdc5RV0v/4ZIdskVCm+73xV23ruWKJ4rH2jhvuEy5vOw9P9nR71xby/ues77HhERAAIAgGEAAgCAYQACAIBhAAIAgGEAAgCAYQACAIBhAAIAgGEAAgCAYQACAIBhAAIAgGEAAgiAmbBZct5BUt+eVIxeP+m+EMmU2SFDP0jsYTpt7RmKG3MSMtFvXP4LLkqf2f5qZySy6dK/vl/72naMgmKxmy3SSpUPSvj2Vt2W6FwQFTfclwPKtzWVNvS75b1Hj/yQ0b1lK2pqRZWo9fb8l2HsYTtp8TMcP980T3UVPvQs4/N9B26MmCAwBMYKYBtGHDBi1cuFB1dXWqq6tTW1ubfv7zn498PpvNqr29XdOnT1dNTY1WrVql7m7/tFUAwMeHaQDNnj1b3/72t7Vz507t2LFD1157rW644Qb99re/lSTde++9euaZZ/Tkk0+qs7NThw8f1k033TQuCwcATG4RZ3lTjVNoaGjQww8/rM9//vOaOXOmNm3apM9//vOSpDfeeEMXX3yxtm3bpk996lNe/fr7+5VOpzXzk/MV9fy7t+U5oLLxOQaeAzplc1P5eD4HZDl5J/NzQBXTpnnXzpzzSVPvY4cOe9fmje9j5IqG4zmOzwEVouP7HJDlx6j1OaBKw8+gk0cOmHr39R/zrrU8B+ScUzFfVl9fn+rq6k5bd8bPAZVKJT3xxBPKZDJqa2vTzp07VSgUtGzZspGaiy66SHPmzNG2bdtO2yeXy6m/v3/UDQAw9ZkH0G9+8xvV1NQolUrpzjvv1ObNm3XJJZeoq6tLyWRS9fX1o+qbmprU1dV12n4dHR1Kp9Mjt9bWVvNGAAAmH/MAWrBggXbt2qVXXnlFd911l1avXq3XX3/9jBewbt069fX1jdwOHjx4xr0AAJOH+XVAyWRS8+fPlyQtXrxY//Vf/6Xvfe97uvnmm5XP59Xb2zvqUVB3d7eam5tP2y+VSimV8n+/ewDA1HDWrwMql8vK5XJavHixEomEtmzZMvK5PXv26MCBA2prazvbbwMAmGJMj4DWrVunlStXas6cORoYGNCmTZu0detWPf/880qn07rtttu0du1aNTQ0qK6uTnfffbfa2tq8r4ADAHx8mAbQ0aNH9ed//uc6cuSI0um0Fi5cqOeff15/8id/Ikn67ne/q2g0qlWrVimXy2n58uX64Q9/eEYLSyQrvC8P9r1cW5ISxsuZLZc/u4jx8mRLrfUqUsMlrc75x4hIkkq2y5lLhkury4ZLnyWpWCh41+bzeVPvYcOl1aXhIVPvovFy5mrDfqlMT7etJe+/DwtZ2z60RvdYRCy9jZf3l4z3N2d4QUC18WUMmf6T3rX9/b2m3pbXMUSjhsgzOUkffa6c9euAxtp7rwNqufCycRlA1tfTTJgBZOosRQ1rMf+MiNiGhOW1VwygU6ue3uhd2zJ/gan3QM8J79pcxradE2UAFTVxBlBdVaWp9/BAn3ft8SO2i7ic87//xAyvW3TOKZ/Lj9/rgAAAOBsMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQDCAAQBAMIABAEAwgAEAQ5jTs8fZeMEPZEvdiedWyNfjBUO8itldbj2cSgiWVYbyTEJwhFsh03I31ZeM7olrendU54zu5Gs9DS5pEqWiLVhrPfWg/uSyt/XuXjUkI1jdntSQhmM9xwz63nleW+jOp/aivmXBRPIcOHeJN6QBgCjh48KBmz5592s9PuAFULpd1+PBh1dbWjvoNp7+/X62trTp48OCHZgtNdmzn1PFx2EaJ7ZxqxmI7nXMaGBhQS0vLh+ZpTrg/wUWj0Q+dmHV1dVP64L+H7Zw6Pg7bKLGdU83Zbmc6nf7IGi5CAAAEwQACAAQxaQZQKpXS/fffr1QqFXop44rtnDo+DtsosZ1Tzbnczgl3EQIA4ONh0jwCAgBMLQwgAEAQDCAAQBAMIABAEJNmAK1fv16f/OQnVVFRoSVLlug///M/Qy9pTH3rW99SJBIZdbvoootCL+usvPzyy7r++uvV0tKiSCSip556atTnnXO67777NGvWLFVWVmrZsmV68803wyz2LHzUdt56660fOLYrVqwIs9gz1NHRoSuuuEK1tbVqbGzUjTfeqD179oyqyWazam9v1/Tp01VTU6NVq1apu7s70IrPjM92XnPNNR84nnfeeWegFZ+ZDRs2aOHChSMvNm1ra9PPf/7zkc+fq2M5KQbQT37yE61du1b333+/fv3rX2vRokVavny5jh49GnppY+rSSy/VkSNHRm6//OUvQy/prGQyGS1atEjr168/5ecfeughff/739ejjz6qV155RdXV1Vq+fLmy2ew5XunZ+ajtlKQVK1aMOraPP/74OVzh2evs7FR7e7u2b9+uF154QYVCQdddd50ymcxIzb333qtnnnlGTz75pDo7O3X48GHddNNNAVdt57OdknT77bePOp4PPfRQoBWfmdmzZ+vb3/62du7cqR07dujaa6/VDTfcoN/+9reSzuGxdJPAlVde6drb20f+XyqVXEtLi+vo6Ai4qrF1//33u0WLFoVexriR5DZv3jzy/3K57Jqbm93DDz888rHe3l6XSqXc448/HmCFY+P92+mcc6tXr3Y33HBDkPWMl6NHjzpJrrOz0zn37rFLJBLuySefHKn53e9+5yS5bdu2hVrmWXv/djrn3B//8R+7v/qrvwq3qHEybdo094//+I/n9FhO+EdA+XxeO3fu1LJly0Y+Fo1GtWzZMm3bti3gysbem2++qZaWFs2bN09f+tKXdODAgdBLGjf79+9XV1fXqOOaTqe1ZMmSKXdcJWnr1q1qbGzUggULdNddd6mnpyf0ks5KX1+fJKmhoUGStHPnThUKhVHH86KLLtKcOXMm9fF8/3a+58c//rFmzJihyy67TOvWrdPQ0FCI5Y2JUqmkJ554QplMRm1tbef0WE64MNL3O378uEqlkpqamkZ9vKmpSW+88UagVY29JUuWaOPGjVqwYIGOHDmiBx54QJ/5zGf02muvqba2NvTyxlxXV5cknfK4vve5qWLFihW66aabNHfuXO3bt09/+7d/q5UrV2rbtm2KxWKhl2dWLpd1zz336KqrrtJll10m6d3jmUwmVV9fP6p2Mh/PU22nJH3xi1/Ueeedp5aWFu3evVtf+9rXtGfPHv3sZz8LuFq73/zmN2pra1M2m1VNTY02b96sSy65RLt27Tpnx3LCD6CPi5UrV478e+HChVqyZInOO+88/fSnP9Vtt90WcGU4W7fccsvIvy+//HItXLhQ559/vrZu3aqlS5cGXNmZaW9v12uvvTbpn6P8KKfbzjvuuGPk35dffrlmzZqlpUuXat++fTr//PPP9TLP2IIFC7Rr1y719fXpX//1X7V69Wp1dnae0zVM+D/BzZgxQ7FY7ANXYHR3d6u5uTnQqsZffX29LrzwQu3duzf0UsbFe8fu43ZcJWnevHmaMWPGpDy2a9as0bPPPqtf/OIXo942pbm5Wfl8Xr29vaPqJ+vxPN12nsqSJUskadIdz2Qyqfnz52vx4sXq6OjQokWL9L3vfe+cHssJP4CSyaQWL16sLVu2jHysXC5ry5YtamtrC7iy8TU4OKh9+/Zp1qxZoZcyLubOnavm5uZRx7W/v1+vvPLKlD6u0rvv+tvT0zOpjq1zTmvWrNHmzZv10ksvae7cuaM+v3jxYiUSiVHHc8+ePTpw4MCkOp4ftZ2nsmvXLkmaVMfzVMrlsnK53Lk9lmN6ScM4eeKJJ1wqlXIbN250r7/+urvjjjtcfX296+rqCr20MfPXf/3XbuvWrW7//v3uV7/6lVu2bJmbMWOGO3r0aOilnbGBgQH36quvuldffdVJct/5znfcq6++6t5++23nnHPf/va3XX19vXv66afd7t273Q033ODmzp3rhoeHA6/c5sO2c2BgwH3lK19x27Ztc/v373cvvvii+8M//EN3wQUXuGw2G3rp3u666y6XTqfd1q1b3ZEjR0ZuQ0NDIzV33nmnmzNnjnvppZfcjh07XFtbm2trawu4aruP2s69e/e6Bx980O3YscPt37/fPf30027evHnu6quvDrxym69//euus7PT7d+/3+3evdt9/etfd5FIxP37v/+7c+7cHctJMYCcc+4HP/iBmzNnjksmk+7KK69027dvD72kMXXzzTe7WbNmuWQy6T7xiU+4m2++2e3duzf0ss7KL37xCyfpA7fVq1c75969FPub3/yma2pqcqlUyi1dutTt2bMn7KLPwIdt59DQkLvuuuvczJkzXSKRcOedd567/fbbJ90vT6faPknuscceG6kZHh52f/mXf+mmTZvmqqqq3Oc+9zl35MiRcIs+Ax+1nQcOHHBXX321a2hocKlUys2fP9/9zd/8jevr6wu7cKO/+Iu/cOedd55LJpNu5syZbunSpSPDx7lzdyx5OwYAQBAT/jkgAMDUxAACAATBAAIABMEAAgAEwQACAATBAAIABMEAAgAEwQACAATBAAIABMEAAgAEwQACAATBAAIABPH/A00yxvCVov0DAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_test_final[2].reshape(32, 32, 3))\n",
    "plt.show()\n",
    "plt.imshow(predict[2].reshape(32, 32, 3))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a98dfb2",
   "metadata": {
    "id": "OQ7532uI3Z9f"
   },
   "source": [
    "The NN model does not success in image comression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "922cdc4e",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "gpuType": "T4",
   "provenance": []
  },
  "gpuClass": "standard",
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}