xavier-initialization-notebook.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "xavier-initialization-notebook.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "3tFGgFKexKF1",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import tensorflow as tf\n",
        "import warnings\n",
        "from matplotlib.colors import ListedColormap\n",
        "from mlxtend.plotting import plot_decision_regions\n",
        "from pylab import rcParams\n",
        "from sklearn.datasets import make_circles, make_moons\n",
        "from sklearn.model_selection import train_test_split\n",
        "from tensorflow.keras.layers import Dense\n",
        "from tensorflow.keras.models import Sequential"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7Le4qZUnxW3P",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "np.set_printoptions(suppress=True)\n",
        "zero_one_colourmap = ListedColormap(('red', 'green'))\n",
        "rcParams['figure.figsize'] = 14, 7\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "# rcParams['axes.facecolor'] = '#383838'"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Iqbfjtwwtq23",
        "colab_type": "text"
      },
      "source": [
        "# Generate Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Hy_Dg1iMyN96",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "X, y = make_circles(n_samples = 1000, \n",
        "                    factor=0.85, \n",
        "                    random_state=2021, \n",
        "                    noise=0.1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XVkM-Ed7xmPs",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "plt.scatter(X[:,0],X[:,1], \n",
        "            c=y, s=100, \n",
        "            cmap = zero_one_colourmap)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lWOSx8a3zDWo",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(X, y, \n",
        "                                                    test_size=0.33, \n",
        "                                                    random_state=42)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VBLshcLzxoeo",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def get_model(inp_activation):\n",
        "  model = Sequential()\n",
        "  model.add(Dense(10,input_dim=2, activation=inp_activation))\n",
        "  model.add(Dense(32, activation = inp_activation))\n",
        "  model.add(Dense(32, activation = inp_activation))\n",
        "  model.add(Dense(1, activation=\"sigmoid\"))\n",
        "  model.compile(loss='binary_crossentropy',\n",
        "                optimizer='adam',\n",
        "                metrics=['accuracy'])\n",
        "  return model"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CA6ssCn7uHcU",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def plot_performance(history):\n",
        "  plt.plot(history.history['loss'])\n",
        "  plt.plot(history.history['val_loss'])\n",
        "  plt.title('Model Loss')\n",
        "  plt.ylabel('Loss')\n",
        "  plt.xlabel('Epoch')\n",
        "  plt.legend(['Train','Val'],loc='upper right')\n",
        "  return plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_Fe3e4HQNDQc",
        "colab_type": "text"
      },
      "source": [
        "# Model With Small Weights"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S6jLHXGM6BPf",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model_1 = get_model(\"tanh\")"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "W2hH-hL75XkW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 294
        },
        "outputId": "c36a66dc-68e3-411c-e455-42159b912167"
      },
      "source": [
        "model_1.summary()"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense_4 (Dense)              (None, 10)                30        \n",
            "_________________________________________________________________\n",
            "dense_5 (Dense)              (None, 32)                352       \n",
            "_________________________________________________________________\n",
            "dense_6 (Dense)              (None, 32)                1056      \n",
            "_________________________________________________________________\n",
            "dense_7 (Dense)              (None, 1)                 33        \n",
            "=================================================================\n",
            "Total params: 1,471\n",
            "Trainable params: 1,471\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T_kG12t5t0Mn",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "small_weight = 0.000001"
      ],
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Bb2RT1mH_d2h",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "initial_weights = model_1.get_weights()"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xjYPUbYityqZ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 103
        },
        "outputId": "fb3b1443-4b9f-432b-e036-5e5f52c15ae0"
      },
      "source": [
        "initial_weights[0]"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[-0.63877743, -0.2595585 , -0.52668774,  0.064161  , -0.30480972,\n",
              "         0.47654957, -0.6302331 , -0.38143656, -0.64366984,  0.18003935],\n",
              "       [-0.00312173, -0.30368337, -0.4492129 ,  0.14035982, -0.58792305,\n",
              "         0.5466513 , -0.13889879,  0.02431417, -0.26729262, -0.06883001]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iPK9jQkk_jmL",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "initial_weights[0] = small_weight * np.ones(initial_weights[0].shape)"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L5Q7HZnzxI1p",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 86
        },
        "outputId": "e4243a73-d2c9-4377-f8a8-646ef32bcb92"
      },
      "source": [
        "initial_weights[0]"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.000001, 0.000001, 0.000001, 0.000001, 0.000001, 0.000001,\n",
              "        0.000001, 0.000001, 0.000001, 0.000001],\n",
              "       [0.000001, 0.000001, 0.000001, 0.000001, 0.000001, 0.000001,\n",
              "        0.000001, 0.000001, 0.000001, 0.000001]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k8_R16S__2JJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "initial_weights[2] = small_weight * np.ones(initial_weights[2].shape)\n",
        "initial_weights[4] = small_weight * np.ones(initial_weights[4].shape)\n",
        "initial_weights[6] = small_weight * np.ones(initial_weights[6].shape)"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7g-tu9SAAcPQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model_1.set_weights(initial_weights)"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bZxYcfAEAcdS",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model_1_history = model_1.fit(X_train, y_train, epochs = 200, \n",
        "                              validation_data = (X_test, y_test),\n",
        "                              verbose=0)"
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xhve3wDNxMJs",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 394
        },
        "outputId": "d2692368-4633-49c2-c1c8-0fd88017df93"
      },
      "source": [
        "plot_performance(model_1_history)"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x504 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NX-FFOB4xMG-",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 381
        },
        "outputId": "5ae52a14-2e0f-4a0d-c384-dc4402600427"
      },
      "source": [
        "plot_decision_regions(X_test, y_test.ravel(), clf = model_1, legend=2)\n",
        "plt.show()"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x504 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SoLBmt5nuQvt",
        "colab_type": "text"
      },
      "source": [
        "# Model With Large Weights"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ILLxZkekA3-4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model_2 = get_model(\"tanh\")"
      ],
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YYz-YOSAA-yh",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "initial_weights_2 = model_2.get_weights()"
      ],
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EMv9OPSdA4Eg",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "initial_weights_2[0] = 50 * np.ones(initial_weights_2[0].shape)\n",
        "initial_weights_2[2] = 50 * np.ones(initial_weights_2[2].shape)\n",
        "initial_weights_2[4] = 50 * np.ones(initial_weights_2[4].shape)\n",
        "initial_weights_2[6] = 50 * np.ones(initial_weights_2[6].shape)"
      ],
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6PFiPNsYBMYV",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model_2.set_weights(initial_weights_2)"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CCeBljuQBMh2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model_2_history = model_2.fit(X_train, y_train, epochs = 200, \n",
        "                              validation_data = (X_test, y_test),\n",
        "                              verbose=0)"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TOBLQMQRBVdd",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 403
        },
        "outputId": "71c210b5-3e05-4bfc-fd0f-35e24c07bca1"
      },
      "source": [
        "plot_performance(model_2_history)"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x504 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7rExhOxZ_S7f",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 381
        },
        "outputId": "abcd067a-d052-4814-f7fd-9921fe6f7b31"
      },
      "source": [
        "plot_decision_regions(X_test, y_test.ravel(), clf = model_2, legend=2)\n",
        "plt.show()"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x504 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "arUK1ssEwTNH",
        "colab_type": "text"
      },
      "source": [
        "# Model With Xavier Initialization"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NOf_boZcCujL",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model = Sequential()\n",
        "model.add(Dense(10,input_dim=2, activation=\"tanh\", \n",
        "                kernel_initializer='glorot_normal',\n",
        "                bias_initializer='zeros'))\n",
        "model.add(Dense(32, activation = \"tanh\",\n",
        "                kernel_initializer='glorot_normal',\n",
        "                bias_initializer='zeros'))\n",
        "model.add(Dense(32, activation = \"tanh\",\n",
        "                kernel_initializer='glorot_normal',\n",
        "                bias_initializer='zeros'))\n",
        "model.add(Dense(1, activation=\"sigmoid\",\n",
        "                kernel_initializer='glorot_normal',\n",
        "                bias_initializer='zeros'))\n",
        "model.compile(loss='binary_crossentropy',\n",
        "              optimizer='adam',\n",
        "              metrics=['accuracy'])"
      ],
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d08kOzIoCuo0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "xavier_weights = model.get_weights()"
      ],
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zwrzru1YCugi",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "8d145efd-f7e4-44dd-cd1a-fbbd6b8c0c63"
      },
      "source": [
        "model_history = model.fit(X_train, y_train, epochs = 200, \n",
        "                          validation_data = (X_test, y_test))"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/200\n",
            "21/21 [==============================] - 0s 11ms/step - loss: 0.7031 - accuracy: 0.4597 - val_loss: 0.6930 - val_accuracy: 0.4909\n",
            "Epoch 2/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6954 - accuracy: 0.4731 - val_loss: 0.6929 - val_accuracy: 0.4788\n",
            "Epoch 3/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6953 - accuracy: 0.4970 - val_loss: 0.6948 - val_accuracy: 0.4576\n",
            "Epoch 4/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6943 - accuracy: 0.4672 - val_loss: 0.6936 - val_accuracy: 0.4424\n",
            "Epoch 5/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6942 - accuracy: 0.4896 - val_loss: 0.6925 - val_accuracy: 0.5273\n",
            "Epoch 6/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6971 - accuracy: 0.4940 - val_loss: 0.6936 - val_accuracy: 0.4939\n",
            "Epoch 7/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6942 - accuracy: 0.4746 - val_loss: 0.6936 - val_accuracy: 0.4182\n",
            "Epoch 8/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6948 - accuracy: 0.4985 - val_loss: 0.6936 - val_accuracy: 0.4879\n",
            "Epoch 9/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6963 - accuracy: 0.4478 - val_loss: 0.6919 - val_accuracy: 0.5091\n",
            "Epoch 10/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6929 - accuracy: 0.5134 - val_loss: 0.6924 - val_accuracy: 0.5091\n",
            "Epoch 11/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6935 - accuracy: 0.4776 - val_loss: 0.6930 - val_accuracy: 0.4758\n",
            "Epoch 12/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6941 - accuracy: 0.5000 - val_loss: 0.6937 - val_accuracy: 0.4758\n",
            "Epoch 13/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6927 - accuracy: 0.5149 - val_loss: 0.6929 - val_accuracy: 0.4970\n",
            "Epoch 14/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6936 - accuracy: 0.4955 - val_loss: 0.6914 - val_accuracy: 0.5152\n",
            "Epoch 15/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6927 - accuracy: 0.4731 - val_loss: 0.6914 - val_accuracy: 0.5303\n",
            "Epoch 16/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.6928 - accuracy: 0.5045 - val_loss: 0.6905 - val_accuracy: 0.5606\n",
            "Epoch 17/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6921 - accuracy: 0.5224 - val_loss: 0.6913 - val_accuracy: 0.5091\n",
            "Epoch 18/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6941 - accuracy: 0.4985 - val_loss: 0.6913 - val_accuracy: 0.5364\n",
            "Epoch 19/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6929 - accuracy: 0.4985 - val_loss: 0.6915 - val_accuracy: 0.5333\n",
            "Epoch 20/200\n",
            "21/21 [==============================] - 0s 5ms/step - loss: 0.6919 - accuracy: 0.5060 - val_loss: 0.6911 - val_accuracy: 0.5606\n",
            "Epoch 21/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6931 - accuracy: 0.5000 - val_loss: 0.6910 - val_accuracy: 0.5485\n",
            "Epoch 22/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6927 - accuracy: 0.5015 - val_loss: 0.6896 - val_accuracy: 0.5455\n",
            "Epoch 23/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6914 - accuracy: 0.5299 - val_loss: 0.6892 - val_accuracy: 0.6000\n",
            "Epoch 24/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.6915 - accuracy: 0.5254 - val_loss: 0.6899 - val_accuracy: 0.5182\n",
            "Epoch 25/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6912 - accuracy: 0.5194 - val_loss: 0.6899 - val_accuracy: 0.5636\n",
            "Epoch 26/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.6910 - accuracy: 0.5224 - val_loss: 0.6886 - val_accuracy: 0.5576\n",
            "Epoch 27/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6918 - accuracy: 0.5269 - val_loss: 0.6872 - val_accuracy: 0.6182\n",
            "Epoch 28/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.6909 - accuracy: 0.5403 - val_loss: 0.6883 - val_accuracy: 0.5515\n",
            "Epoch 29/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6908 - accuracy: 0.5239 - val_loss: 0.6886 - val_accuracy: 0.5515\n",
            "Epoch 30/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6913 - accuracy: 0.5134 - val_loss: 0.6863 - val_accuracy: 0.6061\n",
            "Epoch 31/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6913 - accuracy: 0.5060 - val_loss: 0.6872 - val_accuracy: 0.5788\n",
            "Epoch 32/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6887 - accuracy: 0.5672 - val_loss: 0.6849 - val_accuracy: 0.6242\n",
            "Epoch 33/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6884 - accuracy: 0.5612 - val_loss: 0.6852 - val_accuracy: 0.5788\n",
            "Epoch 34/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6881 - accuracy: 0.5478 - val_loss: 0.6842 - val_accuracy: 0.6182\n",
            "Epoch 35/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6890 - accuracy: 0.5134 - val_loss: 0.6861 - val_accuracy: 0.5576\n",
            "Epoch 36/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.6882 - accuracy: 0.5403 - val_loss: 0.6829 - val_accuracy: 0.6030\n",
            "Epoch 37/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6874 - accuracy: 0.5716 - val_loss: 0.6833 - val_accuracy: 0.5939\n",
            "Epoch 38/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6873 - accuracy: 0.5463 - val_loss: 0.6816 - val_accuracy: 0.6091\n",
            "Epoch 39/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6869 - accuracy: 0.5537 - val_loss: 0.6833 - val_accuracy: 0.6000\n",
            "Epoch 40/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6855 - accuracy: 0.5522 - val_loss: 0.6818 - val_accuracy: 0.5909\n",
            "Epoch 41/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6868 - accuracy: 0.5493 - val_loss: 0.6793 - val_accuracy: 0.6273\n",
            "Epoch 42/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6853 - accuracy: 0.5687 - val_loss: 0.6787 - val_accuracy: 0.5909\n",
            "Epoch 43/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6850 - accuracy: 0.5448 - val_loss: 0.6773 - val_accuracy: 0.6273\n",
            "Epoch 44/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6846 - accuracy: 0.5478 - val_loss: 0.6805 - val_accuracy: 0.5818\n",
            "Epoch 45/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6856 - accuracy: 0.5522 - val_loss: 0.6754 - val_accuracy: 0.6333\n",
            "Epoch 46/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6841 - accuracy: 0.5672 - val_loss: 0.6777 - val_accuracy: 0.5970\n",
            "Epoch 47/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6889 - accuracy: 0.5328 - val_loss: 0.6755 - val_accuracy: 0.6242\n",
            "Epoch 48/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6817 - accuracy: 0.5612 - val_loss: 0.6768 - val_accuracy: 0.6061\n",
            "Epoch 49/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6816 - accuracy: 0.5597 - val_loss: 0.6723 - val_accuracy: 0.6182\n",
            "Epoch 50/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6798 - accuracy: 0.5716 - val_loss: 0.6721 - val_accuracy: 0.6212\n",
            "Epoch 51/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6813 - accuracy: 0.5776 - val_loss: 0.6719 - val_accuracy: 0.6333\n",
            "Epoch 52/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6795 - accuracy: 0.5866 - val_loss: 0.6709 - val_accuracy: 0.6091\n",
            "Epoch 53/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6776 - accuracy: 0.5716 - val_loss: 0.6685 - val_accuracy: 0.6485\n",
            "Epoch 54/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6758 - accuracy: 0.6045 - val_loss: 0.6680 - val_accuracy: 0.6576\n",
            "Epoch 55/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6764 - accuracy: 0.5955 - val_loss: 0.6669 - val_accuracy: 0.6667\n",
            "Epoch 56/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6794 - accuracy: 0.5478 - val_loss: 0.6662 - val_accuracy: 0.6545\n",
            "Epoch 57/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6745 - accuracy: 0.5985 - val_loss: 0.6652 - val_accuracy: 0.6727\n",
            "Epoch 58/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6745 - accuracy: 0.5746 - val_loss: 0.6603 - val_accuracy: 0.6939\n",
            "Epoch 59/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.6698 - accuracy: 0.6463 - val_loss: 0.6598 - val_accuracy: 0.7030\n",
            "Epoch 60/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6660 - accuracy: 0.6836 - val_loss: 0.6564 - val_accuracy: 0.7061\n",
            "Epoch 61/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6658 - accuracy: 0.6627 - val_loss: 0.6560 - val_accuracy: 0.7424\n",
            "Epoch 62/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6613 - accuracy: 0.7254 - val_loss: 0.6522 - val_accuracy: 0.7485\n",
            "Epoch 63/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.6588 - accuracy: 0.7119 - val_loss: 0.6519 - val_accuracy: 0.7364\n",
            "Epoch 64/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6571 - accuracy: 0.6761 - val_loss: 0.6467 - val_accuracy: 0.7152\n",
            "Epoch 65/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6554 - accuracy: 0.6791 - val_loss: 0.6450 - val_accuracy: 0.7333\n",
            "Epoch 66/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6548 - accuracy: 0.6746 - val_loss: 0.6451 - val_accuracy: 0.7152\n",
            "Epoch 67/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6479 - accuracy: 0.6791 - val_loss: 0.6380 - val_accuracy: 0.7576\n",
            "Epoch 68/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6468 - accuracy: 0.6552 - val_loss: 0.6327 - val_accuracy: 0.7424\n",
            "Epoch 69/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6420 - accuracy: 0.6612 - val_loss: 0.6346 - val_accuracy: 0.7152\n",
            "Epoch 70/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6426 - accuracy: 0.6612 - val_loss: 0.6251 - val_accuracy: 0.7333\n",
            "Epoch 71/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6364 - accuracy: 0.7254 - val_loss: 0.6218 - val_accuracy: 0.7788\n",
            "Epoch 72/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6320 - accuracy: 0.7164 - val_loss: 0.6211 - val_accuracy: 0.7424\n",
            "Epoch 73/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6263 - accuracy: 0.7134 - val_loss: 0.6162 - val_accuracy: 0.7364\n",
            "Epoch 74/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6260 - accuracy: 0.7000 - val_loss: 0.6121 - val_accuracy: 0.7030\n",
            "Epoch 75/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6272 - accuracy: 0.6731 - val_loss: 0.6125 - val_accuracy: 0.7061\n",
            "Epoch 76/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6173 - accuracy: 0.7030 - val_loss: 0.6026 - val_accuracy: 0.7364\n",
            "Epoch 77/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6135 - accuracy: 0.6851 - val_loss: 0.6051 - val_accuracy: 0.7091\n",
            "Epoch 78/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6091 - accuracy: 0.7358 - val_loss: 0.5927 - val_accuracy: 0.7667\n",
            "Epoch 79/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6053 - accuracy: 0.7239 - val_loss: 0.5889 - val_accuracy: 0.7576\n",
            "Epoch 80/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.6027 - accuracy: 0.7313 - val_loss: 0.5876 - val_accuracy: 0.7364\n",
            "Epoch 81/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5923 - accuracy: 0.7522 - val_loss: 0.5748 - val_accuracy: 0.7818\n",
            "Epoch 82/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5896 - accuracy: 0.7448 - val_loss: 0.5829 - val_accuracy: 0.6970\n",
            "Epoch 83/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5851 - accuracy: 0.7269 - val_loss: 0.5738 - val_accuracy: 0.7242\n",
            "Epoch 84/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5778 - accuracy: 0.7493 - val_loss: 0.5625 - val_accuracy: 0.7485\n",
            "Epoch 85/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5708 - accuracy: 0.7881 - val_loss: 0.5526 - val_accuracy: 0.7879\n",
            "Epoch 86/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5690 - accuracy: 0.7507 - val_loss: 0.5486 - val_accuracy: 0.7758\n",
            "Epoch 87/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5611 - accuracy: 0.7567 - val_loss: 0.5594 - val_accuracy: 0.7485\n",
            "Epoch 88/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5662 - accuracy: 0.7388 - val_loss: 0.5431 - val_accuracy: 0.7636\n",
            "Epoch 89/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.5484 - accuracy: 0.7910 - val_loss: 0.5435 - val_accuracy: 0.7424\n",
            "Epoch 90/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5561 - accuracy: 0.7522 - val_loss: 0.5300 - val_accuracy: 0.7758\n",
            "Epoch 91/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5433 - accuracy: 0.7418 - val_loss: 0.5489 - val_accuracy: 0.6970\n",
            "Epoch 92/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5497 - accuracy: 0.7448 - val_loss: 0.5205 - val_accuracy: 0.7667\n",
            "Epoch 93/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5373 - accuracy: 0.7657 - val_loss: 0.5290 - val_accuracy: 0.7545\n",
            "Epoch 94/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5382 - accuracy: 0.7582 - val_loss: 0.5194 - val_accuracy: 0.7424\n",
            "Epoch 95/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5350 - accuracy: 0.7731 - val_loss: 0.5224 - val_accuracy: 0.7364\n",
            "Epoch 96/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5244 - accuracy: 0.7731 - val_loss: 0.5018 - val_accuracy: 0.7879\n",
            "Epoch 97/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5210 - accuracy: 0.7761 - val_loss: 0.5027 - val_accuracy: 0.7727\n",
            "Epoch 98/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5173 - accuracy: 0.7851 - val_loss: 0.5015 - val_accuracy: 0.7485\n",
            "Epoch 99/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5199 - accuracy: 0.7701 - val_loss: 0.4991 - val_accuracy: 0.7485\n",
            "Epoch 100/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5263 - accuracy: 0.7478 - val_loss: 0.5105 - val_accuracy: 0.7303\n",
            "Epoch 101/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5234 - accuracy: 0.7567 - val_loss: 0.4882 - val_accuracy: 0.7879\n",
            "Epoch 102/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5096 - accuracy: 0.7701 - val_loss: 0.4934 - val_accuracy: 0.7758\n",
            "Epoch 103/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5099 - accuracy: 0.7627 - val_loss: 0.4856 - val_accuracy: 0.7848\n",
            "Epoch 104/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5115 - accuracy: 0.7687 - val_loss: 0.4829 - val_accuracy: 0.7879\n",
            "Epoch 105/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5047 - accuracy: 0.7672 - val_loss: 0.4847 - val_accuracy: 0.7697\n",
            "Epoch 106/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5004 - accuracy: 0.7746 - val_loss: 0.4880 - val_accuracy: 0.7697\n",
            "Epoch 107/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4978 - accuracy: 0.7821 - val_loss: 0.4771 - val_accuracy: 0.7970\n",
            "Epoch 108/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4953 - accuracy: 0.7910 - val_loss: 0.4808 - val_accuracy: 0.7545\n",
            "Epoch 109/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5069 - accuracy: 0.7552 - val_loss: 0.5067 - val_accuracy: 0.7091\n",
            "Epoch 110/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.5019 - accuracy: 0.7612 - val_loss: 0.5088 - val_accuracy: 0.7273\n",
            "Epoch 111/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4951 - accuracy: 0.7776 - val_loss: 0.4850 - val_accuracy: 0.7424\n",
            "Epoch 112/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4871 - accuracy: 0.7672 - val_loss: 0.4728 - val_accuracy: 0.7606\n",
            "Epoch 113/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4910 - accuracy: 0.7851 - val_loss: 0.4771 - val_accuracy: 0.7606\n",
            "Epoch 114/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4832 - accuracy: 0.7821 - val_loss: 0.4670 - val_accuracy: 0.7848\n",
            "Epoch 115/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4880 - accuracy: 0.7657 - val_loss: 0.4783 - val_accuracy: 0.7545\n",
            "Epoch 116/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4928 - accuracy: 0.7701 - val_loss: 0.4891 - val_accuracy: 0.7424\n",
            "Epoch 117/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4843 - accuracy: 0.7836 - val_loss: 0.4650 - val_accuracy: 0.7636\n",
            "Epoch 118/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4852 - accuracy: 0.7776 - val_loss: 0.4638 - val_accuracy: 0.7788\n",
            "Epoch 119/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4857 - accuracy: 0.7716 - val_loss: 0.4724 - val_accuracy: 0.7485\n",
            "Epoch 120/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4787 - accuracy: 0.7866 - val_loss: 0.4663 - val_accuracy: 0.7576\n",
            "Epoch 121/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4814 - accuracy: 0.7716 - val_loss: 0.4685 - val_accuracy: 0.7727\n",
            "Epoch 122/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4933 - accuracy: 0.7612 - val_loss: 0.4659 - val_accuracy: 0.7515\n",
            "Epoch 123/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4795 - accuracy: 0.7866 - val_loss: 0.4632 - val_accuracy: 0.7727\n",
            "Epoch 124/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4704 - accuracy: 0.7940 - val_loss: 0.4634 - val_accuracy: 0.7727\n",
            "Epoch 125/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4763 - accuracy: 0.7731 - val_loss: 0.4780 - val_accuracy: 0.7667\n",
            "Epoch 126/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.4809 - accuracy: 0.7746 - val_loss: 0.4707 - val_accuracy: 0.7515\n",
            "Epoch 127/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4832 - accuracy: 0.7746 - val_loss: 0.4616 - val_accuracy: 0.7667\n",
            "Epoch 128/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4779 - accuracy: 0.7881 - val_loss: 0.4582 - val_accuracy: 0.7788\n",
            "Epoch 129/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4739 - accuracy: 0.7761 - val_loss: 0.4690 - val_accuracy: 0.7455\n",
            "Epoch 130/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4701 - accuracy: 0.7896 - val_loss: 0.4584 - val_accuracy: 0.7606\n",
            "Epoch 131/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.4775 - accuracy: 0.7836 - val_loss: 0.4895 - val_accuracy: 0.7667\n",
            "Epoch 132/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4914 - accuracy: 0.7642 - val_loss: 0.4845 - val_accuracy: 0.7485\n",
            "Epoch 133/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4806 - accuracy: 0.7642 - val_loss: 0.4590 - val_accuracy: 0.7758\n",
            "Epoch 134/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4748 - accuracy: 0.7896 - val_loss: 0.4593 - val_accuracy: 0.7636\n",
            "Epoch 135/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4708 - accuracy: 0.7836 - val_loss: 0.4684 - val_accuracy: 0.7576\n",
            "Epoch 136/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4846 - accuracy: 0.7806 - val_loss: 0.4661 - val_accuracy: 0.7788\n",
            "Epoch 137/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4745 - accuracy: 0.7731 - val_loss: 0.4556 - val_accuracy: 0.7606\n",
            "Epoch 138/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4658 - accuracy: 0.7836 - val_loss: 0.4641 - val_accuracy: 0.7697\n",
            "Epoch 139/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4692 - accuracy: 0.7896 - val_loss: 0.4750 - val_accuracy: 0.7394\n",
            "Epoch 140/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4760 - accuracy: 0.7746 - val_loss: 0.4621 - val_accuracy: 0.7667\n",
            "Epoch 141/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4664 - accuracy: 0.7910 - val_loss: 0.4555 - val_accuracy: 0.7697\n",
            "Epoch 142/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4710 - accuracy: 0.7642 - val_loss: 0.4859 - val_accuracy: 0.7606\n",
            "Epoch 143/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4732 - accuracy: 0.7866 - val_loss: 0.4641 - val_accuracy: 0.7485\n",
            "Epoch 144/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4662 - accuracy: 0.7866 - val_loss: 0.4598 - val_accuracy: 0.7727\n",
            "Epoch 145/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4748 - accuracy: 0.7761 - val_loss: 0.4542 - val_accuracy: 0.7788\n",
            "Epoch 146/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4688 - accuracy: 0.7821 - val_loss: 0.4591 - val_accuracy: 0.7606\n",
            "Epoch 147/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4760 - accuracy: 0.7657 - val_loss: 0.4593 - val_accuracy: 0.7727\n",
            "Epoch 148/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4697 - accuracy: 0.7701 - val_loss: 0.4643 - val_accuracy: 0.7667\n",
            "Epoch 149/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4656 - accuracy: 0.7731 - val_loss: 0.4778 - val_accuracy: 0.7515\n",
            "Epoch 150/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4838 - accuracy: 0.7821 - val_loss: 0.4619 - val_accuracy: 0.7697\n",
            "Epoch 151/200\n",
            "21/21 [==============================] - 0s 5ms/step - loss: 0.4650 - accuracy: 0.7925 - val_loss: 0.4555 - val_accuracy: 0.7818\n",
            "Epoch 152/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4710 - accuracy: 0.7746 - val_loss: 0.4772 - val_accuracy: 0.7606\n",
            "Epoch 153/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4751 - accuracy: 0.7761 - val_loss: 0.4733 - val_accuracy: 0.7576\n",
            "Epoch 154/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4701 - accuracy: 0.7836 - val_loss: 0.4709 - val_accuracy: 0.7788\n",
            "Epoch 155/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4743 - accuracy: 0.7687 - val_loss: 0.4608 - val_accuracy: 0.7545\n",
            "Epoch 156/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4771 - accuracy: 0.7701 - val_loss: 0.4573 - val_accuracy: 0.7879\n",
            "Epoch 157/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4737 - accuracy: 0.7716 - val_loss: 0.4644 - val_accuracy: 0.7697\n",
            "Epoch 158/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4641 - accuracy: 0.7925 - val_loss: 0.4628 - val_accuracy: 0.7606\n",
            "Epoch 159/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4652 - accuracy: 0.7866 - val_loss: 0.4646 - val_accuracy: 0.7727\n",
            "Epoch 160/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4660 - accuracy: 0.7806 - val_loss: 0.4552 - val_accuracy: 0.7879\n",
            "Epoch 161/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4697 - accuracy: 0.7776 - val_loss: 0.4700 - val_accuracy: 0.7576\n",
            "Epoch 162/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4673 - accuracy: 0.7746 - val_loss: 0.4555 - val_accuracy: 0.7667\n",
            "Epoch 163/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4696 - accuracy: 0.7866 - val_loss: 0.4953 - val_accuracy: 0.7636\n",
            "Epoch 164/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4801 - accuracy: 0.7731 - val_loss: 0.4711 - val_accuracy: 0.7515\n",
            "Epoch 165/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4642 - accuracy: 0.7866 - val_loss: 0.4564 - val_accuracy: 0.7727\n",
            "Epoch 166/200\n",
            "21/21 [==============================] - 0s 5ms/step - loss: 0.4725 - accuracy: 0.7761 - val_loss: 0.4695 - val_accuracy: 0.7455\n",
            "Epoch 167/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4762 - accuracy: 0.7716 - val_loss: 0.4732 - val_accuracy: 0.7576\n",
            "Epoch 168/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4676 - accuracy: 0.7851 - val_loss: 0.4637 - val_accuracy: 0.7515\n",
            "Epoch 169/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4612 - accuracy: 0.7985 - val_loss: 0.4648 - val_accuracy: 0.7576\n",
            "Epoch 170/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4652 - accuracy: 0.7881 - val_loss: 0.4618 - val_accuracy: 0.7667\n",
            "Epoch 171/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4669 - accuracy: 0.7761 - val_loss: 0.4595 - val_accuracy: 0.7606\n",
            "Epoch 172/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4679 - accuracy: 0.7716 - val_loss: 0.4658 - val_accuracy: 0.7636\n",
            "Epoch 173/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4670 - accuracy: 0.7806 - val_loss: 0.4596 - val_accuracy: 0.7697\n",
            "Epoch 174/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4714 - accuracy: 0.7731 - val_loss: 0.4689 - val_accuracy: 0.7576\n",
            "Epoch 175/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4629 - accuracy: 0.7821 - val_loss: 0.4624 - val_accuracy: 0.7788\n",
            "Epoch 176/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4670 - accuracy: 0.7851 - val_loss: 0.4577 - val_accuracy: 0.7727\n",
            "Epoch 177/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4751 - accuracy: 0.7821 - val_loss: 0.4773 - val_accuracy: 0.7515\n",
            "Epoch 178/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4588 - accuracy: 0.7866 - val_loss: 0.4603 - val_accuracy: 0.7758\n",
            "Epoch 179/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4708 - accuracy: 0.7731 - val_loss: 0.4814 - val_accuracy: 0.7515\n",
            "Epoch 180/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.4676 - accuracy: 0.7672 - val_loss: 0.4611 - val_accuracy: 0.7667\n",
            "Epoch 181/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4625 - accuracy: 0.7791 - val_loss: 0.4652 - val_accuracy: 0.7545\n",
            "Epoch 182/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4646 - accuracy: 0.7851 - val_loss: 0.4694 - val_accuracy: 0.7515\n",
            "Epoch 183/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4629 - accuracy: 0.7776 - val_loss: 0.4558 - val_accuracy: 0.7818\n",
            "Epoch 184/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4701 - accuracy: 0.7821 - val_loss: 0.4893 - val_accuracy: 0.7485\n",
            "Epoch 185/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4623 - accuracy: 0.7731 - val_loss: 0.4595 - val_accuracy: 0.7939\n",
            "Epoch 186/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4626 - accuracy: 0.7940 - val_loss: 0.4588 - val_accuracy: 0.7727\n",
            "Epoch 187/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4615 - accuracy: 0.7955 - val_loss: 0.4574 - val_accuracy: 0.7727\n",
            "Epoch 188/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4644 - accuracy: 0.7761 - val_loss: 0.4727 - val_accuracy: 0.7485\n",
            "Epoch 189/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4631 - accuracy: 0.7791 - val_loss: 0.4642 - val_accuracy: 0.7636\n",
            "Epoch 190/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4607 - accuracy: 0.7836 - val_loss: 0.4635 - val_accuracy: 0.7667\n",
            "Epoch 191/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4643 - accuracy: 0.7821 - val_loss: 0.4571 - val_accuracy: 0.7697\n",
            "Epoch 192/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4575 - accuracy: 0.7970 - val_loss: 0.4609 - val_accuracy: 0.7697\n",
            "Epoch 193/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4659 - accuracy: 0.7746 - val_loss: 0.4806 - val_accuracy: 0.7485\n",
            "Epoch 194/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4642 - accuracy: 0.7866 - val_loss: 0.4650 - val_accuracy: 0.7758\n",
            "Epoch 195/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4648 - accuracy: 0.7806 - val_loss: 0.4665 - val_accuracy: 0.7576\n",
            "Epoch 196/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4691 - accuracy: 0.7791 - val_loss: 0.4656 - val_accuracy: 0.7667\n",
            "Epoch 197/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4620 - accuracy: 0.7746 - val_loss: 0.4562 - val_accuracy: 0.7818\n",
            "Epoch 198/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4690 - accuracy: 0.7970 - val_loss: 0.4781 - val_accuracy: 0.7455\n",
            "Epoch 199/200\n",
            "21/21 [==============================] - 0s 6ms/step - loss: 0.4632 - accuracy: 0.7896 - val_loss: 0.4610 - val_accuracy: 0.7788\n",
            "Epoch 200/200\n",
            "21/21 [==============================] - 0s 7ms/step - loss: 0.4594 - accuracy: 0.7776 - val_loss: 0.4724 - val_accuracy: 0.7515\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yKKPHDVzBVg8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 402
        },
        "outputId": "70e5808c-f17e-4de1-f42f-8d5a2900a00a"
      },
      "source": [
        "plot_performance(model_history)"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x504 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ysLhIrOpAnkQ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 381
        },
        "outputId": "95dbae86-1392-4b2b-ccc3-525691dae06e"
      },
      "source": [
        "plot_decision_regions(X_test, y_test.ravel(), clf = model, legend=2)\n",
        "plt.show()"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x504 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MkFQVOOz-XGy",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}