TF_Forum_15324.ipynb

tf_forum_15324.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyMWBXpEE+XOU90IOzmfOipS",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/kiransair/0eaa42a92df29a52a64b400b04b1055a/tf_forum_15324.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zC3wpNs0QFcO",
        "outputId": "fa3cd1e3-1391-4d0a-aea0-525ffa1763cc"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2.11.0\n"
          ]
        }
      ],
      "source": [
        "# TensorFlow and tf.keras\n",
        "import tensorflow as tf\n",
        "\n",
        "# Helper libraries\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "print(tf.__version__)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fashion_mnist = tf.keras.datasets.fashion_mnist\n",
        "\n",
        "(train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "z2mv2uirQUTw",
        "outputId": "030171db-374a-455e-d65a-27c80b6539f5"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-labels-idx1-ubyte.gz\n",
            "29515/29515 [==============================] - 0s 0us/step\n",
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-images-idx3-ubyte.gz\n",
            "26421880/26421880 [==============================] - 0s 0us/step\n",
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-labels-idx1-ubyte.gz\n",
            "5148/5148 [==============================] - 0s 0us/step\n",
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-images-idx3-ubyte.gz\n",
            "4422102/4422102 [==============================] - 0s 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "class_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat',\n",
        "               'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle boot']"
      ],
      "metadata": {
        "id": "GnkdBOElQWJj"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure()\n",
        "plt.imshow(train_images[0])\n",
        "plt.colorbar()\n",
        "plt.grid(False)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "bPGHQ24cQaR5",
        "outputId": "d7d4bea6-6922-4caa-cc19-c672a0b3fd15"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_images = train_images / 255.0\n",
        "\n",
        "test_images = test_images / 255.0"
      ],
      "metadata": {
        "id": "pJKeXJarQddK"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(10,10))\n",
        "for i in range(25):\n",
        "    plt.subplot(5,5,i+1)\n",
        "    plt.xticks([])\n",
        "    plt.yticks([])\n",
        "    plt.grid(False)\n",
        "    plt.imshow(train_images[i], cmap=plt.cm.binary)\n",
        "    plt.xlabel(class_names[train_labels[i]])\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 592
        },
        "id": "n0hMl_75QgF2",
        "outputId": "81aa5b75-568a-447a-c92d-ceee888250cc"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x720 with 25 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model = tf.keras.Sequential([\n",
        "    tf.keras.layers.Flatten(input_shape=(28, 28)),\n",
        "    tf.keras.layers.Dense(128, activation='relu'),\n",
        "    tf.keras.layers.Dense(10)\n",
        "])"
      ],
      "metadata": {
        "id": "8lhv0_G_QiCb"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(optimizer='adam',\n",
        "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
        "              metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "vivjNn1-QlWi"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.fit(train_images, train_labels, epochs=10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uY1XYgl3QnnR",
        "outputId": "8139e88d-da89-4231-af24-74c5725a5fc7"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "1875/1875 [==============================] - 10s 4ms/step - loss: 0.4997 - accuracy: 0.8234\n",
            "Epoch 2/10\n",
            "1875/1875 [==============================] - 8s 4ms/step - loss: 0.3763 - accuracy: 0.8652\n",
            "Epoch 3/10\n",
            "1875/1875 [==============================] - 8s 4ms/step - loss: 0.3377 - accuracy: 0.8772\n",
            "Epoch 4/10\n",
            "1875/1875 [==============================] - 10s 5ms/step - loss: 0.3139 - accuracy: 0.8833\n",
            "Epoch 5/10\n",
            "1875/1875 [==============================] - 9s 5ms/step - loss: 0.2950 - accuracy: 0.8916\n",
            "Epoch 6/10\n",
            "1875/1875 [==============================] - 11s 6ms/step - loss: 0.2805 - accuracy: 0.8953\n",
            "Epoch 7/10\n",
            "1875/1875 [==============================] - 7s 4ms/step - loss: 0.2693 - accuracy: 0.9007\n",
            "Epoch 8/10\n",
            "1875/1875 [==============================] - 10s 5ms/step - loss: 0.2569 - accuracy: 0.9043\n",
            "Epoch 9/10\n",
            "1875/1875 [==============================] - 10s 5ms/step - loss: 0.2473 - accuracy: 0.9068\n",
            "Epoch 10/10\n",
            "1875/1875 [==============================] - 8s 4ms/step - loss: 0.2381 - accuracy: 0.9110\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fbb6146a5e0>"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)\n",
        "\n",
        "print('\\nTest accuracy:', test_acc)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "T5uPPVZrQptt",
        "outputId": "c7a0f279-9747-453b-ee16-4ca2bb9885cc"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 - 1s - loss: 0.3599 - accuracy: 0.8752 - 1s/epoch - 5ms/step\n",
            "\n",
            "Test accuracy: 0.8751999735832214\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "probability_model = tf.keras.Sequential([model, \n",
        "                                         tf.keras.layers.Softmax()])"
      ],
      "metadata": {
        "id": "CBIf2RHLQu9C"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "predictions = probability_model.predict(test_images)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_QIlz6xTQxjj",
        "outputId": "a80746d0-e40c-4ab8-c36f-40b2f220a2b1"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 [==============================] - 1s 3ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictions[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Valx-8meQzXm",
        "outputId": "c248bce9-981a-4b94-a217-5a1d85414ffa"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([2.1128129e-07, 2.0696269e-09, 2.6192307e-08, 7.0714199e-09,\n",
              "       1.0259638e-08, 1.8476669e-03, 2.6627836e-07, 2.1214534e-02,\n",
              "       3.1344185e-09, 9.7693729e-01], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "np.argmax(predictions[0])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QNTEQj-XQ1Sc",
        "outputId": "e6363a1a-c43a-4d3c-c589-c4c5fac29f24"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "9"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_labels[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fBuUPnb7Q5f4",
        "outputId": "a6fa71ec-82fc-4ec3-a5bd-28a65ecf8133"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "9"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_image(i, predictions_array, true_label, img):\n",
        "  true_label, img = true_label[i], img[i]\n",
        "  plt.grid(False)\n",
        "  plt.xticks([])\n",
        "  plt.yticks([])\n",
        "\n",
        "  plt.imshow(img, cmap=plt.cm.binary)\n",
        "\n",
        "  predicted_label = np.argmax(predictions_array)\n",
        "  if predicted_label == true_label:\n",
        "    color = 'blue'\n",
        "  else:\n",
        "    color = 'red'\n",
        "\n",
        "  plt.xlabel(\"{} {:2.0f}% ({})\".format(class_names[predicted_label],\n",
        "                                100*np.max(predictions_array),\n",
        "                                class_names[true_label]),\n",
        "                                color=color)\n",
        "\n",
        "def plot_value_array(i, predictions_array, true_label):\n",
        "  true_label = true_label[i]\n",
        "  plt.grid(False)\n",
        "  plt.xticks(range(10))\n",
        "  plt.yticks([])\n",
        "  thisplot = plt.bar(range(10), predictions_array, color=\"#777777\")\n",
        "  plt.ylim([0, 1])\n",
        "  predicted_label = np.argmax(predictions_array)\n",
        "\n",
        "  thisplot[predicted_label].set_color('red')\n",
        "  thisplot[true_label].set_color('blue')"
      ],
      "metadata": {
        "id": "2p1vmKcJQ8QC"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "i = 0\n",
        "\n",
        "plt.figure(figsize=(6,3))\n",
        "\n",
        "plt.subplot(1,2,1)\n",
        "\n",
        "plot_image(i, predictions[i], test_labels, test_images)\n",
        "\n",
        "plt.subplot(1,2,2)\n",
        "\n",
        "plot_value_array(i, predictions[i], test_labels)\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 211
        },
        "id": "3XUNVPsVRDXc",
        "outputId": "e08442b0-248f-43c0-bc2c-6df8b79df7cb"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x216 with 2 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAADCCAYAAAB3whgdAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAAsTAAALEwEAmpwYAAAS10lEQVR4nO3de7BdVX3A8e8vCZAQAoTwToCrbSJVoUEyVIsyIkgFEcHiaOqjVsfaogU61oJ9YLVja1unlalaWnlaIeqEUFErISoiQYhJeCUBEiwkEcIrIEkIkBBY/WPv4M3dayf7JDesPL6fmTu553fXOmvtnXN/d5211t4nUkpIkl5+Q0p3QJJ2ViZgSSrEBCxJhZiAJakQE7AkFWIClqRChpXugFTavvvum/r6+kp3QzuouXPnLk8p7Zf7mQlYO72+vj7mzJlTuhvaQUXEkrafOQUhSYWYgCUp48ADIaL714EH9t6GCViSMh59dOuWhx7ngF2s0Na0ePFili9fHqX7Ib1cekrALlZoa5o0aVLpLkgvK6cgJKkQE7AkFWIClqRCTMCSVIgJWJIKMQFLUiEmYEkqxAQsSYWYgCWpEBOwJBViApakQkzAklSICViSCjEBS1IhJmBJKsQELEmFmIAlqRATsCQVYgKWpEJMwJJUiAlYkgoxAUtSISZgSSrEBCxJhZiAJakQE7AkFWIClqRCTMCSVIgJWJIKMQFLUiEmYEkqxAQsSYWYgCWpEBOwJBViApakQoaV7oAGzwsvvNCIDRmS/xsbEZ2fd82aNY3Ybrvtli173333NWLjx4/v3Ja0M3EELEmFmIAlqRATsCQVYgKWpEJMwJJUiLsgBklKqVMM8jsTHnrooWzZW265pRE7+eSTs2VHjhy5sS5utrYdDznTpk1rxM4777zB7I60w3AELEmFmIAlqRATsCQVYgKWpEJchNuK2i4Dzrnpppuy8VmzZjViy5Yty5Y9++yzO7fXi8cee6wRmz59erbsqFGjtkofpB2RI2BJKsQELEmFmIAlqRATsCQVYgKWpELcBTFIcjdDHzYsf3pnz57diN1zzz3ZsgcccEAjlrvpOcAZZ5zRiI0ePTpb9rnnnmvEDjvssGzZJ554ohFbuXJltuzYsWOzcUlNjoAlqRATsCQVYgKWpEJMwJJUiItwm+HFF19sxHILbqtXr87Wnzp1aiPWds/d3GLZqlWrsmV7uSdxLr5gwYJs2XHjxjVibYt7ucVISXmOgCWpEBOwJBViApakQkzAklSICViSCtnudkHkVu8jIls2t1uhrWwu3raiP3To0I118SUXXXRRNp67vHj48OHZskuWLGnEcjsj2p533bp12bK54237VOXcDo0VK1Zky65Zs6YRa9sNsrU+xVnaXjgClqRCTMCSVIgJWJIKMQFLUiHbxCJcLwtrbfGcXj6VOLfg1nWxDWDKlCmN2COPPJIte9RRRzVibYtlTz31VCO2zz77ZMuOGTOmEVu+fHm27NNPP925Dzltlzg/88wzjVjb/YsnTpzYuT1pR+QIWJIKMQFLUiEmYEkqxAQsSYVsE4twvSys5a5uy8Ugv4jW1lYvC26XXnppI7Zo0aJG7JBDDsnWz33IZdui1rPPPtuItX3wZe4+wW3Hu/vuuzdibVfY9bJImjN9+vRs3EU47ewcAUtSISZgSSrEBCxJhZiAJakQE7AkFbLVdkG07UzIya2ot+0KyF1e3Mslx22WLVvWiE2bNi1bNrczYfz48Y1Y7nJfyN8zN7czAmCXXXZpxNp2IOQuA26TO2dtn8ycK9t2L99c326++ebO/ZJ2Jo6AJakQE7AkFWIClqRCTMCSVEjPi3AD75vbdgnvli6M9XKp6+OPP56NL168uBFbuHBhtuzDDz/ciO26667ZsnvuuWcjlrtv78qVK7P1n3/++UYstzAH+fObOy7I38937733zpbNHVvbh5DmFkRHjBiRLZt7jj322CNbdv78+Rs8zi1uSjsyR8CSVIgJWJIKMQFLUiEmYEkqxAQsSYX0vAui643LH3300UZsyZIl2bKrV6/uFIP8SvkDDzyQLZu7NHfYsPwhjxo1qhFru5x6xYoVnfrV1lauX227CnKXB69duzZb9qCDDmrE2nZi5PowevTobNncJdVPPvlktmxux0Pbp0MPfI62XRjSjsoRsCQVYgKWpEJMwJJUiAlYkgrZ4vsB//CHP8zGc/fXbVuUyl1K3LYgk1sE7GVhre0evbmForZ7EucuG84tYLUt4uX60Ha8ufvutl3am7vsuO0y7V7kjq3tUvPcYmTbomHb/5u0s3AELEmFmIAlqRATsCQVYgKWpEJMwJJUSE/L0CtXruT666/fIHbJJZdkyx5++OGNWO5SWejtMuAtvZF4ri3Ir9S3rfSvWrWqU1ttNxjP3Wy+7RhyuzNyl3kD3H333Y1Y2w6EXi77ze26aLtUfPjw4Z3qA+y///4bPM59ArS0I3MELEmFmIAlqRATsCQVYgKWpEJ6WoQbOXIkxxxzzAaxW2+9NVt23rx5jdjMmTM7t9W2IJNbRNtnn32yZXPxvfbaK1s2t1jVdinyE0880YjlPm05d89dyN+jt+1ToO+8885G7Mgjj8yW7evra8RmzJiRLZu7nLqXT7Juu4z44IMPbsRynyINzcVM7wesnY0jYEkqxAQsSYWYgCWpEBOwJBViApakQnraBTF06NDGTb8vuOCCzvXbboY+a9asRiy3qwDgZz/7WSO2ePHibNm77rqrEWu7hDa346FtZ0Jut0Bux8URRxyRrX/iiSc2Yqecckq2bO7S3l6cdtpp2fjSpUsbsTFjxmTL5nYxtF3SndsdkftkZ4AJEyZs8HhLj1Xa3jgClqRCTMCSVIgJWJIKMQFLUiEv68fStt0X9oQTTugUAzjrrLMGtU87umuvvbZ0Fzrr5VJoaUfgK16SCjEBS1IhJmBJKsQELEmFmIAlqRATsCQVYgKWpEJMwJJUiAlYkgoxAUtSISZgSSrEBCxJhZiAJakQE7AkFWIClqRCTMCSVIgJWJIKMQFLUiEmYEkqxAQsSYWYgCWpEBOwJBViApakQkzAklSICViSCjEBS1IhJmBJKsQELEmFmIAlqRATsCQVYgKWpEJMwJJUyLBeCs+dO3d5RCzZWp3RTu+w0h2QXk49JeCU0n5bqyOStLPpKQFLUinnn39+57Jf+MIXtmJPBo9zwJJUiAlYkgpxCkLSZtkRpwRebpFSKt0HqaiIeBzodXfPvsDyzWjOettOmy9XvcPaNjCYgKXNEBFzUkqTrDc49Uq0WeIYB3IOWJIKGbQEHMHpEaQIDu9YfnEE+2biT/fYbk/lN/I8H4rg4Jaf/XYEt0QwL4LvRrBnHd8lgivq+D0RfLqO7xfBzAjmR3B6v+f5Tlsb9c/PjeCD/R4Pi+DxCDpNoEXwdxH8RSZ+eQRndnmOzSm/keeZGMEp/R6fGsHntvR5pR3FYI6AJwMz63+3Rx+C1uR4MXB+ShwBXAN8qo6/G9itjh8NfCyCPqpzcBFwDHAuQATvAG5PiWW5BiIYBnwYuKpf+K3AIuDdEcTmHlhBE+HXCRj4PvCOCHYv051B9V/WG9R6JdoscYwbGJQEHMEewBuBjwDv7Rd/cwQ/iWBqBPdGcOXARBLBiAh+EMFHM8/7qQhmR3BXBJ/dSPv/FsGCCH4UwX51bGIEt9Z1r4lgdFu8Hu1NAq6M4I4IRgxoYgLw0/r7GcDv198nYGSdPEcAa4GVwPPA7sBuwAv1z88F/nkjp/EtwG0psa5fbDJwIbAUeEO/410cwWcjuK0efTfedUTw0fq8jhgQPzqCGyOYG8H0CA5q6c+JEcyJYFEEp9Z1h0dwWd3m7REc3xaPYFfgc8B76nP6npRIwE+ger7tWUpps34JrbfttFniGAcarBHwO4HrUmIR8EQER/f72VFUyefVwCuBY/v9bA/gu8CUlPha/yeM4CRgPNUociJwdATHZdoeCcxJidcANwKfqeNfB85LiSOBeRuLp8RUYA7wvpSYmBLPDmhjQX2MUI16D6m/nwqsBh6mSpJfTIknqUax76RK1v8AnAX8d0o8k+n/escCc/sd/3DgxPXnh+Y7i+Up8TrgP2DDaYcIPkGV5E7vfywR7AL8O3BmShwNXAp8vqU/fVTn/u3ARXV/Pg6kesQ/GbiiLU712roA+FZ9Tr9VP+8c4E0bOQ/STmOwEvBk4Jv1999kw2Tx85R4MCVeBO6g+sVe7zvAZSnx9cxznlR/3Q7cBhxOlZAHehFe+uX+BvDGCPYC9k6JG+v4FcBxbfEOx/dh4KwI5gKjqEa6UCWoF6imLl4BfDKCV6bEipR4e0pMqvv+DmBqBF+r3w28IdPGQcDj/R6fCtxQJ9CrgdMjGNrv59Pqf+ey4Tn9IHAyVZJdM6CNVwGvBWZEcAfwN8C4lmP+dkq8mBL3AfdTnf83Up1jUuJeqq1bEzYSz3mM9qmebV5EvC0iFkbELyKi80bYiLg0Ih6LiPk9tndIRNwQEXdHxIKIOKdjveER8fOIuLOu1/oOsqX+0Ii4PSK+10OdxRExLyLuiIg5PdTbOyKmRsS9EXFPROR+PwbWeVXdzvqvlRFxbsf2/rw+J/MjYkpEDO9Y75y6zoKubW3KFl+IEcE+VG+fj4ggAUOBFPHSPGn/JPDCgDZvBt4WwVX129MNnhr4x5T4zx67NOj76uqkchJABBOoRoUAf0A18n8eeCyCm6mmMu7vV/1vqUaZ6+fIp1Ilz98b0MyzQP8XwmSqPyaL68djqM7zjPrx+vM68JzOo3rHMA54YEAbASxIKfsHYKCB53GwzutwaLzD2C5ExFDgK1Rz8w8CsyPi2pTS3R2qXw58GbKDjY1ZB3wypXRbRIwC5kbEjA5trgHeklJ6OiJ2AWZGxA9SSrd2bPcc4B6oFpx7cHxKqde9tRcC16WUzoyIXWHTawQppYVUr/P1/y8PUa3PbFREjAXOBl6dUno2Ir5NNW16+SbqvRb4KNWgay1wXUR8L6X0i021uTGDMQI+k+rt9WEp0ZcSh1D94nd5m3kB8CuqF/VA04EP1/PLRDA2gv0z5YbUfYAqIc5MiRXAryJe6sMHgBvb4vX3q6hGtw3r241gCNWo8aL6R0upkiIRjAReD9zbr954YFxK/ITqRfUiVSIbOMcM1Yv9N+t6e1Kdv0Prc9pH9Ta/ywLn7cDHgGszOy4WAvutH4FHtYvjNS3P8+4IhkTwG1RTRwuBm4D31XUnAIduIp47pxOAnkaB25BjgF+klO5PKa2lerf3zk3UASCl9FPgyV4bTCk9nFK6rf5+FdXrZGyHeimltH6H0C71V6c/ohExjmqQcXGv/e1VROxF9S70EoCU0tqU0lM9Ps0JwP+llLpeTDMMGBERw6h+L7ML4wP8FjArpfRMSmkdVd54V4/9bBiMBDyZ5l+eq+m+G+IcYETEhgtUKXE91VzqLRHMoxo55hLkauCYCOZTJcP125z+EPiXCO6i+ku5qfjlVHOduUW4yREsokquy4DL6vhXgD0iWADMpppOuatfvc8Df11/PwX407rchZnj+AG/ng45A/jxgCmE71DtINgtU3cDKTGTal74+9Fvq19KrKX6Y/VPEdxJNSX0uy1PsxT4ed2vP0mJ54CvAkPq/49vAR+q+9gWvwF49fpFuPp5j6faDbE9Ggv8st/jB+mQDAdLRPRRranM6lh+aETcQTXtMyOl1Kke8CXgL6kGDL1IwPURMTci/rhjnVdQTb1dVk95XBwRI3ts971Uv1+b7mBKDwFfpHp9PwysSCld36HqfOBNETEmInan2t1zyCbqdOqQX9vIF6RrII0v3Y+teHwHQPpR6X5sfv85E7i43+MPAF/uoX4fMH8z296Dar7/XZtRd2+qP4av7VD2VOCr9fdvBr7XQztj63/3B+4EjutQZxLVNMvv1I8vBP6+hzZ3pbos+ICO5UcDPwb2o3pX8D/A+zvW/Uj9f/BTqsXvL23pa8or4bYt50PrtrAdwaHAJ0t3Ygs8xIajnnF1bKuq53CvBq5MKU3bVPmBUvWW/gbgbR2KHwucFhGLqaZY3hIR3+jYzkP1v49RvSs+pkO1B4EH069H51OB13Vpr3YycFtK6dGO5U8EHkgpPZ5Sep5qPabtXeAGUkqXpJSOTikdRzV1uqiHfmaZgLchKbEwpZf2G+9wUmJ2StxRuh9bYDYwPiJeUS8WvRe4dms2GBFBNT96T0rpX3uot19E7F1/P4Jq4fDejVYCUkqfTimNSyn1UR3fj1NK7+/Q3sh6kZB6CuEkOsz1p5QeAX4ZEa+qQycAXRY115tMx+mH2lLg9RGxe31uT6CaV9+kiKjXguJQqvnfqzZeY9O8HaXUUUppXUR8gmqBeChwaUppQZe6ETGF6i39vhHxIPCZlNIlHaoeSzXVMa+ezwX4q5TS/26i3kHAFfUOgSHAt1NKnbeUbYYDgGuqnMYw4KqU0nUd6/4ZcGX9R+1+4I+6VKoT/VupFp07SSnNioipVNtD11EtWne9sOLqiBhDdaHVx1Pvi4UN3g1NkgpxCkKSCjEBS1IhJmBJKsQELEmFmIAlqRATsCQVYgKWpEJMwJJUyP8DtPMG8VmnFc8AAAAASUVORK5CYII=\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "52VB1OJ_RPH3"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}