Deeplizard.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Deeplizard.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyOU4GE7028vyEcq2KRVes8F",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Santigs/AI/blob/master/Deeplizard.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "687jnAZRXljH",
        "colab_type": "text"
      },
      "source": [
        "# Deep learning exercise from DeepLizard [Videocourse](https://www.youtube.com/watch?v=UkzhouEk6uY&list=PLZbbT5o_s2xrwRnXk_yCPtnqqo4_u2YGL&index=3)\n",
        "\n",
        "---\n",
        "\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IDAefKlqgE8H",
        "colab_type": "text"
      },
      "source": [
        "### Creating a random set of sample data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sWK85TjpZ3bO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "from random import randint\n",
        "from sklearn.preprocessing import MinMaxScaler"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JX1IQflMa56F",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_labels = []\n",
        "train_samples = []"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pgHa4cjVbTBY",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "for i in range(50):\n",
        "  random_younger = randint(13,64)\n",
        "  train_samples.append(random_younger)\n",
        "  train_labels.append(1)\n",
        "\n",
        "  random_older = randint(65,100)\n",
        "  train_samples.append(random_older)\n",
        "  train_labels.append(0)\n",
        "\n",
        "for i in range(1000):\n",
        "  random_younger = randint(13,64)\n",
        "  train_samples.append(random_younger)\n",
        "  train_labels.append(0)\n",
        "\n",
        "  random_older = randint(65,100)\n",
        "  train_samples.append(random_older)\n",
        "  train_labels.append(1)\n",
        "\n",
        "\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UrNFb57bdY8a",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "for i in train_samples:\n",
        "  print(i)\n",
        "  "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Gj6FfZdIeZdo",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "for i in train_labels:\n",
        "  print(i)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vG4vktycdvFZ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_labels = np.array(train_labels)\n",
        "train_samples = np.array(train_samples)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UbSG2cg6EDcW",
        "colab_type": "code",
        "outputId": "455a2c84-9597-4420-83c3-d90ea3ab21d9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "train_samples"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 59,  75,  31, ..., 100,  61,  66])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oOsoOScYEHLP",
        "colab_type": "code",
        "outputId": "d7bc4a25-2388-4458-8607-e0b7af422a68",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "train_labels"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1, 0, 1, ..., 1, 0, 1])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lf1rhDi9duL-",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "scaler = MinMaxScaler(feature_range=(0,1))\n",
        "scaled_train_samples = scaler.fit_transform((train_samples).reshape(-1,1))\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HJ10aia1EM8-",
        "colab_type": "code",
        "outputId": "f0b32e27-2e60-4ab1-dab5-852fa1d20fee",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        }
      },
      "source": [
        "scaled_train_samples"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.52873563],\n",
              "       [0.71264368],\n",
              "       [0.20689655],\n",
              "       ...,\n",
              "       [1.        ],\n",
              "       [0.55172414],\n",
              "       [0.6091954 ]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dm2KIGgyfYH2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "for i in scaled_train_samples:\n",
        "  print(i)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dInWeicHf8AU",
        "colab_type": "text"
      },
      "source": [
        "### Defining the Model Neural Network Architecture"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ILB1dkdJUwJT",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "%tensorflow_version 1.x\n",
        "import keras\n",
        "from keras import backend as K\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Activation\n",
        "from keras.layers.core import Dense\n",
        "from keras.optimizers import Adam\n",
        "from keras.metrics import categorical_crossentropy"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fvBbF0w1U6hB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model = Sequential([\n",
        "                    Dense(16, input_shape=(1,), activation='relu' ),\n",
        "                    Dense(32, activation='relu'),\n",
        "                    Dense(2, activation='softmax')\n",
        "])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "80nPcqh6XQBg",
        "colab_type": "code",
        "outputId": "e88a4382-866e-4d2d-b1dc-51c1a7407c96",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 255
        }
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_3\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense_7 (Dense)              (None, 16)                32        \n",
            "_________________________________________________________________\n",
            "dense_8 (Dense)              (None, 32)                544       \n",
            "_________________________________________________________________\n",
            "dense_9 (Dense)              (None, 2)                 66        \n",
            "=================================================================\n",
            "Total params: 642\n",
            "Trainable params: 642\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Tt7mu1APuxW3",
        "colab_type": "text"
      },
      "source": [
        "### Training the Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i1kSMr-pXSIv",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model.compile(Adam(lr=.0001), loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XnSG637AtXI_",
        "colab_type": "code",
        "outputId": "b9522eb5-d69d-4418-82a1-e8a86dd762f4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 731
        }
      },
      "source": [
        "# make sure the validation set (which is the % at the end of the sample set) is representative of the whole population\n",
        "model.fit(scaled_train_samples, train_labels, validation_split=0.1   , batch_size=10, epochs=20, shuffle=True, verbose=2)"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 1890 samples, validate on 210 samples\n",
            "Epoch 1/20\n",
            " - 0s - loss: 0.6934 - acc: 0.5688 - val_loss: 0.6688 - val_acc: 0.5762\n",
            "Epoch 2/20\n",
            " - 0s - loss: 0.6536 - acc: 0.6090 - val_loss: 0.6296 - val_acc: 0.6667\n",
            "Epoch 3/20\n",
            " - 0s - loss: 0.6197 - acc: 0.6852 - val_loss: 0.5917 - val_acc: 0.7524\n",
            "Epoch 4/20\n",
            " - 0s - loss: 0.5882 - acc: 0.7307 - val_loss: 0.5573 - val_acc: 0.7952\n",
            "Epoch 5/20\n",
            " - 0s - loss: 0.5582 - acc: 0.7741 - val_loss: 0.5231 - val_acc: 0.8286\n",
            "Epoch 6/20\n",
            " - 0s - loss: 0.5289 - acc: 0.8026 - val_loss: 0.4899 - val_acc: 0.8524\n",
            "Epoch 7/20\n",
            " - 0s - loss: 0.5004 - acc: 0.8344 - val_loss: 0.4567 - val_acc: 0.8762\n",
            "Epoch 8/20\n",
            " - 0s - loss: 0.4732 - acc: 0.8476 - val_loss: 0.4255 - val_acc: 0.9095\n",
            "Epoch 9/20\n",
            " - 0s - loss: 0.4480 - acc: 0.8603 - val_loss: 0.3968 - val_acc: 0.9190\n",
            "Epoch 10/20\n",
            " - 0s - loss: 0.4251 - acc: 0.8794 - val_loss: 0.3695 - val_acc: 0.9286\n",
            "Epoch 11/20\n",
            " - 0s - loss: 0.4046 - acc: 0.8841 - val_loss: 0.3452 - val_acc: 0.9286\n",
            "Epoch 12/20\n",
            " - 0s - loss: 0.3864 - acc: 0.8910 - val_loss: 0.3232 - val_acc: 0.9286\n",
            "Epoch 13/20\n",
            " - 0s - loss: 0.3706 - acc: 0.8937 - val_loss: 0.3044 - val_acc: 0.9381\n",
            "Epoch 14/20\n",
            " - 0s - loss: 0.3575 - acc: 0.9079 - val_loss: 0.2870 - val_acc: 0.9381\n",
            "Epoch 15/20\n",
            " - 0s - loss: 0.3459 - acc: 0.9106 - val_loss: 0.2715 - val_acc: 0.9381\n",
            "Epoch 16/20\n",
            " - 0s - loss: 0.3360 - acc: 0.9101 - val_loss: 0.2584 - val_acc: 0.9476\n",
            "Epoch 17/20\n",
            " - 0s - loss: 0.3278 - acc: 0.9185 - val_loss: 0.2463 - val_acc: 0.9381\n",
            "Epoch 18/20\n",
            " - 0s - loss: 0.3208 - acc: 0.9148 - val_loss: 0.2362 - val_acc: 0.9476\n",
            "Epoch 19/20\n",
            " - 0s - loss: 0.3148 - acc: 0.9190 - val_loss: 0.2275 - val_acc: 0.9476\n",
            "Epoch 20/20\n",
            " - 0s - loss: 0.3096 - acc: 0.9228 - val_loss: 0.2195 - val_acc: 0.9476\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7f54fc953780>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UT_2gszr2QCu",
        "colab_type": "text"
      },
      "source": [
        "### Preprocess Test data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OHAb27PT2O-z",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "test_labels = []\n",
        "test_samples = []"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LJbqwI6_2ffq",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "for i in range(10):\n",
        "  random_younger = randint(13,64)\n",
        "  test_samples.append(random_younger)\n",
        "  test_labels.append(1)\n",
        "\n",
        "  random_older = randint(65,100)\n",
        "  test_samples.append(random_older)\n",
        "  test_labels.append(0)\n",
        "\n",
        "for i in range(200):\n",
        "  random_younger = randint(13,64)\n",
        "  test_samples.append(random_younger)\n",
        "  test_labels.append(0)\n",
        "\n",
        "  random_older = randint(65,100)\n",
        "  test_samples.append(random_older)\n",
        "  test_labels.append(1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dpew9VdW2652",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "test_labels = np.array(test_labels)\n",
        "test_samples = np.array(test_samples)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y2hSZeAT3BSL",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "scaler = MinMaxScaler(feature_range=(0,1))\n",
        "scaled_test_samples = scaler.fit_transform((test_samples).reshape(-1,1))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ioxWU7Cj3TRH",
        "colab_type": "text"
      },
      "source": [
        "### Predict"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ATwLxHgE3WIO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "predictions = model.predict(scaled_test_samples, batch_size=10, verbose=0)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zImnHBjz3rw1",
        "colab_type": "code",
        "outputId": "3cecf1b2-b654-4815-c860-97d20f54338a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "predictions"
      ],
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.5221015 , 0.47789845],\n",
              "       [0.04295469, 0.95704526],\n",
              "       [0.94246536, 0.05753459],\n",
              "       [0.05008588, 0.9499141 ],\n",
              "       [0.5903305 , 0.40966955],\n",
              "       [0.12543945, 0.87456053],\n",
              "       [0.91519904, 0.08480096],\n",
              "       [0.06134874, 0.93865126],\n",
              "       [0.91519904, 0.08480096],\n",
              "       [0.05544875, 0.9445513 ],\n",
              "       [0.94258916, 0.05741079],\n",
              "       [0.09049767, 0.9095023 ],\n",
              "       [0.82708263, 0.1729173 ],\n",
              "       [0.05008588, 0.9499141 ],\n",
              "       [0.9423414 , 0.05765864],\n",
              "       [0.07713682, 0.9228631 ],\n",
              "       [0.5678396 , 0.4321604 ],\n",
              "       [0.4078206 , 0.5921795 ],\n",
              "       [0.90832347, 0.09167654],\n",
              "       [0.43028393, 0.5697161 ],\n",
              "       [0.94367784, 0.05632218],\n",
              "       [0.1718259 , 0.8281741 ],\n",
              "       [0.8924949 , 0.10750502],\n",
              "       [0.30270958, 0.6972904 ],\n",
              "       [0.9435705 , 0.0564295 ],\n",
              "       [0.06451436, 0.9354856 ],\n",
              "       [0.9006894 , 0.09931056],\n",
              "       [0.32253662, 0.6774634 ],\n",
              "       [0.9435705 , 0.0564295 ],\n",
              "       [0.36411208, 0.6358879 ],\n",
              "       [0.9436304 , 0.05636967],\n",
              "       [0.4078206 , 0.5921795 ],\n",
              "       [0.83988607, 0.16011398],\n",
              "       [0.18535908, 0.81464094],\n",
              "       [0.9423069 , 0.05769311],\n",
              "       [0.19970109, 0.80029887],\n",
              "       [0.9432044 , 0.05679562],\n",
              "       [0.36411208, 0.6358879 ],\n",
              "       [0.9429591 , 0.05704095],\n",
              "       [0.19970109, 0.80029887],\n",
              "       [0.94258916, 0.05741079],\n",
              "       [0.43028393, 0.5697161 ],\n",
              "       [0.9352644 , 0.06473562],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.9423069 , 0.05769311],\n",
              "       [0.12543945, 0.87456053],\n",
              "       [0.8134829 , 0.1865171 ],\n",
              "       [0.30270958, 0.6972904 ],\n",
              "       [0.82708263, 0.1729173 ],\n",
              "       [0.06451436, 0.9354856 ],\n",
              "       [0.9423069 , 0.05769311],\n",
              "       [0.24762939, 0.7523706 ],\n",
              "       [0.9427128 , 0.05728727],\n",
              "       [0.18535908, 0.81464094],\n",
              "       [0.47598985, 0.5240101 ],\n",
              "       [0.08334444, 0.91665554],\n",
              "       [0.65525043, 0.34474957],\n",
              "       [0.10654986, 0.89345014],\n",
              "       [0.5678396 , 0.4321604 ],\n",
              "       [0.28359118, 0.71640885],\n",
              "       [0.82708263, 0.1729173 ],\n",
              "       [0.06134874, 0.93865126],\n",
              "       [0.8631802 , 0.13681984],\n",
              "       [0.06451436, 0.9354856 ],\n",
              "       [0.9430818 , 0.05691816],\n",
              "       [0.04759225, 0.9524077 ],\n",
              "       [0.94331163, 0.05668839],\n",
              "       [0.08334444, 0.91665554],\n",
              "       [0.94367784, 0.05632217],\n",
              "       [0.32253662, 0.6774634 ],\n",
              "       [0.7838464 , 0.2161536 ],\n",
              "       [0.12543945, 0.87456053],\n",
              "       [0.8519111 , 0.14808889],\n",
              "       [0.08334444, 0.91665554],\n",
              "       [0.7332854 , 0.2667146 ],\n",
              "       [0.19970109, 0.80029887],\n",
              "       [0.5678396 , 0.4321604 ],\n",
              "       [0.0981991 , 0.90180093],\n",
              "       [0.8835553 , 0.11644472],\n",
              "       [0.04521686, 0.95478314],\n",
              "       [0.94331163, 0.05668839],\n",
              "       [0.1471263 , 0.8528737 ],\n",
              "       [0.91519904, 0.08480096],\n",
              "       [0.38573614, 0.6142639 ],\n",
              "       [0.94246536, 0.05753459],\n",
              "       [0.19970109, 0.80029887],\n",
              "       [0.94331163, 0.05668839],\n",
              "       [0.32253662, 0.6774634 ],\n",
              "       [0.942217  , 0.05778298],\n",
              "       [0.13591875, 0.86408126],\n",
              "       [0.8924949 , 0.10750502],\n",
              "       [0.1471263 , 0.8528737 ],\n",
              "       [0.94258916, 0.05741079],\n",
              "       [0.12543945, 0.87456053],\n",
              "       [0.5221015 , 0.47789845],\n",
              "       [0.1718259 , 0.8281741 ],\n",
              "       [0.9433267 , 0.05667334],\n",
              "       [0.12543945, 0.87456053],\n",
              "       [0.942836  , 0.05716397],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.87371886, 0.12628116],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.6757894 , 0.32421055],\n",
              "       [0.21486023, 0.7851398 ],\n",
              "       [0.90832347, 0.09167654],\n",
              "       [0.43028393, 0.5697161 ],\n",
              "       [0.9210364 , 0.0789636 ],\n",
              "       [0.06809378, 0.9319062 ],\n",
              "       [0.79907334, 0.20092666],\n",
              "       [0.36411208, 0.6358879 ],\n",
              "       [0.7148519 , 0.28514808],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.93152344, 0.06847661],\n",
              "       [0.08334444, 0.91665554],\n",
              "       [0.93838054, 0.06161946],\n",
              "       [0.28359118, 0.71640885],\n",
              "       [0.94367784, 0.05632218],\n",
              "       [0.06134874, 0.93865126],\n",
              "       [0.5903305 , 0.40966955],\n",
              "       [0.13591875, 0.86408126],\n",
              "       [0.8835553 , 0.11644472],\n",
              "       [0.06451436, 0.9354856 ],\n",
              "       [0.9430818 , 0.05691816],\n",
              "       [0.05270293, 0.94729704],\n",
              "       [0.8519111 , 0.14808889],\n",
              "       [0.05270293, 0.94729704],\n",
              "       [0.9435705 , 0.0564295 ],\n",
              "       [0.1718259 , 0.8281741 ],\n",
              "       [0.45303786, 0.5469622 ],\n",
              "       [0.38573614, 0.6142639 ],\n",
              "       [0.5678396 , 0.4321604 ],\n",
              "       [0.24762939, 0.7523706 ],\n",
              "       [0.9352644 , 0.06473562],\n",
              "       [0.38573614, 0.6142639 ],\n",
              "       [0.87371886, 0.12628116],\n",
              "       [0.38573614, 0.6142639 ],\n",
              "       [0.79907334, 0.20092666],\n",
              "       [0.05544875, 0.9445513 ],\n",
              "       [0.7509424 , 0.24905756],\n",
              "       [0.3430234 , 0.65697664],\n",
              "       [0.6757894 , 0.32421055],\n",
              "       [0.10654986, 0.89345014],\n",
              "       [0.94064754, 0.05935246],\n",
              "       [0.3430234 , 0.65697664],\n",
              "       [0.942836  , 0.05716397],\n",
              "       [0.43028393, 0.5697161 ],\n",
              "       [0.94246536, 0.05753459],\n",
              "       [0.19970109, 0.80029887],\n",
              "       [0.94258916, 0.05741079],\n",
              "       [0.0981991 , 0.90180093],\n",
              "       [0.6956731 , 0.30432686],\n",
              "       [0.09049767, 0.9095023 ],\n",
              "       [0.7838464 , 0.2161536 ],\n",
              "       [0.265221  , 0.73477906],\n",
              "       [0.5450657 , 0.4549343 ],\n",
              "       [0.06451436, 0.9354856 ],\n",
              "       [0.8631802 , 0.13681984],\n",
              "       [0.1718259 , 0.8281741 ],\n",
              "       [0.94246536, 0.05753459],\n",
              "       [0.07713682, 0.9228631 ],\n",
              "       [0.9432044 , 0.05679562],\n",
              "       [0.4078206 , 0.5921795 ],\n",
              "       [0.94258916, 0.05741079],\n",
              "       [0.1471263 , 0.8528737 ],\n",
              "       [0.5678396 , 0.4321604 ],\n",
              "       [0.05544875, 0.9445513 ],\n",
              "       [0.6757894 , 0.32421055],\n",
              "       [0.1156599 , 0.8843401 ],\n",
              "       [0.47598985, 0.5240101 ],\n",
              "       [0.32253662, 0.6774634 ],\n",
              "       [0.6341147 , 0.3658853 ],\n",
              "       [0.06451436, 0.9354856 ],\n",
              "       [0.65525043, 0.34474957],\n",
              "       [0.0981991 , 0.90180093],\n",
              "       [0.9429591 , 0.05704095],\n",
              "       [0.04295469, 0.95704526],\n",
              "       [0.9430818 , 0.05691816],\n",
              "       [0.09049767, 0.9095023 ],\n",
              "       [0.8924949 , 0.10750502],\n",
              "       [0.12543945, 0.87456053],\n",
              "       [0.9432044 , 0.05679562],\n",
              "       [0.13591875, 0.86408126],\n",
              "       [0.9352644 , 0.06473562],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.8835553 , 0.11644472],\n",
              "       [0.18535908, 0.81464094],\n",
              "       [0.9430818 , 0.05691816],\n",
              "       [0.30270958, 0.6972904 ],\n",
              "       [0.94344866, 0.05655129],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.9436659 , 0.05633406],\n",
              "       [0.05544875, 0.9445513 ],\n",
              "       [0.9436304 , 0.05636967],\n",
              "       [0.06134874, 0.93865126],\n",
              "       [0.90832347, 0.09167654],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.9436659 , 0.05633406],\n",
              "       [0.1156599 , 0.8843401 ],\n",
              "       [0.83988607, 0.16011398],\n",
              "       [0.17182592, 0.8281741 ],\n",
              "       [0.79907334, 0.20092666],\n",
              "       [0.09049767, 0.9095023 ],\n",
              "       [0.9423414 , 0.05765864],\n",
              "       [0.05544875, 0.9445513 ],\n",
              "       [0.9423069 , 0.05769311],\n",
              "       [0.09049767, 0.9095023 ],\n",
              "       [0.5450657 , 0.4549343 ],\n",
              "       [0.0981991 , 0.90180093],\n",
              "       [0.9210364 , 0.07896359],\n",
              "       [0.17182592, 0.8281741 ],\n",
              "       [0.9433267 , 0.05667334],\n",
              "       [0.05008588, 0.9499141 ],\n",
              "       [0.9433267 , 0.05667334],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.7148519 , 0.28514808],\n",
              "       [0.32253662, 0.6774634 ],\n",
              "       [0.7148519 , 0.28514808],\n",
              "       [0.21486023, 0.7851398 ],\n",
              "       [0.9432044 , 0.05679562],\n",
              "       [0.30270958, 0.6972904 ],\n",
              "       [0.8519111 , 0.14808889],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.8519111 , 0.14808889],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.5221015 , 0.47789845],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.8519111 , 0.14808889],\n",
              "       [0.06134874, 0.93865126],\n",
              "       [0.49904364, 0.5009564 ],\n",
              "       [0.32253662, 0.6774634 ],\n",
              "       [0.90832347, 0.09167654],\n",
              "       [0.08334444, 0.91665554],\n",
              "       [0.9210364 , 0.0789636 ],\n",
              "       [0.09049767, 0.9095023 ],\n",
              "       [0.9429591 , 0.05704095],\n",
              "       [0.28359118, 0.71640885],\n",
              "       [0.94344866, 0.05655129],\n",
              "       [0.1471263 , 0.8528737 ],\n",
              "       [0.90832347, 0.09167654],\n",
              "       [0.24762939, 0.7523706 ],\n",
              "       [0.8519111 , 0.14808889],\n",
              "       [0.1718259 , 0.8281741 ],\n",
              "       [0.94367784, 0.05632218],\n",
              "       [0.1718259 , 0.8281741 ],\n",
              "       [0.91519904, 0.08480096],\n",
              "       [0.08334444, 0.91665554],\n",
              "       [0.94367784, 0.05632218],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.9423069 , 0.05769311],\n",
              "       [0.0981991 , 0.90180093],\n",
              "       [0.45303786, 0.5469622 ],\n",
              "       [0.06451436, 0.9354856 ],\n",
              "       [0.7678007 , 0.23219931],\n",
              "       [0.05008588, 0.9499141 ],\n",
              "       [0.9429591 , 0.05704095],\n",
              "       [0.0981991 , 0.90180093],\n",
              "       [0.7838464 , 0.2161536 ],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.9436304 , 0.05636967],\n",
              "       [0.04295469, 0.95704526],\n",
              "       [0.93152344, 0.06847661],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.7678007 , 0.23219931],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.82708263, 0.1729173 ],\n",
              "       [0.07713682, 0.9228631 ],\n",
              "       [0.94344866, 0.05655129],\n",
              "       [0.04759225, 0.9524077 ],\n",
              "       [0.9433267 , 0.05667334],\n",
              "       [0.04521686, 0.95478314],\n",
              "       [0.49904364, 0.5009564 ],\n",
              "       [0.15908778, 0.8409122 ],\n",
              "       [0.5678396 , 0.4321604 ],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.8519111 , 0.14808889],\n",
              "       [0.04759225, 0.9524077 ],\n",
              "       [0.9430818 , 0.05691816],\n",
              "       [0.36411208, 0.6358879 ],\n",
              "       [0.45303786, 0.5469622 ],\n",
              "       [0.18535908, 0.81464094],\n",
              "       [0.9423414 , 0.05765864],\n",
              "       [0.0720837 , 0.9279163 ],\n",
              "       [0.6956731 , 0.30432686],\n",
              "       [0.4078206 , 0.5921795 ],\n",
              "       [0.91519904, 0.08480096],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.9423069 , 0.05769311],\n",
              "       [0.05544875, 0.9445513 ],\n",
              "       [0.93152344, 0.06847661],\n",
              "       [0.17182592, 0.8281741 ],\n",
              "       [0.7838464 , 0.2161536 ],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.942217  , 0.05778298],\n",
              "       [0.08334444, 0.91665554],\n",
              "       [0.8631802 , 0.13681984],\n",
              "       [0.07713682, 0.9228631 ],\n",
              "       [0.94344866, 0.05655129],\n",
              "       [0.265221  , 0.73477906],\n",
              "       [0.9006894 , 0.09931055],\n",
              "       [0.28359118, 0.71640885],\n",
              "       [0.5450657 , 0.4549343 ],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.93152344, 0.06847661],\n",
              "       [0.06451436, 0.9354856 ],\n",
              "       [0.6956731 , 0.30432686],\n",
              "       [0.265221  , 0.73477906],\n",
              "       [0.93152344, 0.06847661],\n",
              "       [0.0720837 , 0.9279163 ],\n",
              "       [0.7678007 , 0.23219931],\n",
              "       [0.13591875, 0.86408126],\n",
              "       [0.94246536, 0.05753459],\n",
              "       [0.21486023, 0.7851398 ],\n",
              "       [0.5450657 , 0.4549343 ],\n",
              "       [0.05008588, 0.9499141 ],\n",
              "       [0.942217  , 0.05778298],\n",
              "       [0.36411208, 0.6358879 ],\n",
              "       [0.90832347, 0.09167654],\n",
              "       [0.43028393, 0.5697161 ],\n",
              "       [0.79907334, 0.20092666],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.8631802 , 0.13681984],\n",
              "       [0.04759225, 0.9524077 ],\n",
              "       [0.7678007 , 0.23219931],\n",
              "       [0.15908778, 0.8409122 ],\n",
              "       [0.65525043, 0.34474957],\n",
              "       [0.06134874, 0.93865126],\n",
              "       [0.65525043, 0.34474957],\n",
              "       [0.06451436, 0.9354856 ],\n",
              "       [0.7148519 , 0.28514808],\n",
              "       [0.04759225, 0.9524077 ],\n",
              "       [0.9423414 , 0.05765864],\n",
              "       [0.06134874, 0.93865126],\n",
              "       [0.9433267 , 0.05667334],\n",
              "       [0.30270958, 0.6972904 ],\n",
              "       [0.8519111 , 0.14808889],\n",
              "       [0.21486023, 0.7851398 ],\n",
              "       [0.91519904, 0.08480096],\n",
              "       [0.30270958, 0.6972904 ],\n",
              "       [0.9427128 , 0.05728726],\n",
              "       [0.05832881, 0.94167113],\n",
              "       [0.91519904, 0.08480096],\n",
              "       [0.30270958, 0.6972904 ],\n",
              "       [0.93838054, 0.06161946],\n",
              "       [0.13591875, 0.86408126],\n",
              "       [0.8134829 , 0.1865171 ],\n",
              "       [0.4078206 , 0.5921795 ],\n",
              "       [0.942836  , 0.05716397],\n",
              "       [0.43028393, 0.5697161 ],\n",
              "       [0.47598985, 0.5240101 ],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.7332854 , 0.2667146 ],\n",
              "       [0.10654986, 0.89345014],\n",
              "       [0.9423414 , 0.05765864],\n",
              "       [0.05270293, 0.94729704],\n",
              "       [0.7509424 , 0.24905755],\n",
              "       [0.06134874, 0.93865126],\n",
              "       [0.9423414 , 0.05765864],\n",
              "       [0.05008588, 0.9499141 ],\n",
              "       [0.65525043, 0.34474957],\n",
              "       [0.08334444, 0.91665554],\n",
              "       [0.93152344, 0.06847661],\n",
              "       [0.36411208, 0.6358879 ],\n",
              "       [0.9430818 , 0.05691816],\n",
              "       [0.06451436, 0.9354856 ],\n",
              "       [0.6956731 , 0.30432686],\n",
              "       [0.04521686, 0.95478314],\n",
              "       [0.9423069 , 0.05769311],\n",
              "       [0.06451436, 0.9354856 ],\n",
              "       [0.942836  , 0.05716397],\n",
              "       [0.0981991 , 0.90180093],\n",
              "       [0.49904364, 0.5009564 ],\n",
              "       [0.19970109, 0.80029887],\n",
              "       [0.6341147 , 0.3658853 ],\n",
              "       [0.05270293, 0.94729704],\n",
              "       [0.94367784, 0.05632218],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.93152344, 0.06847661],\n",
              "       [0.43028393, 0.5697161 ],\n",
              "       [0.6341147 , 0.3658853 ],\n",
              "       [0.04521686, 0.95478314],\n",
              "       [0.87371886, 0.12628116],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.90832347, 0.09167654],\n",
              "       [0.0720837 , 0.9279163 ],\n",
              "       [0.7678007 , 0.23219931],\n",
              "       [0.05008588, 0.9499141 ],\n",
              "       [0.83988607, 0.16011398],\n",
              "       [0.23083808, 0.76916194],\n",
              "       [0.9210364 , 0.07896359],\n",
              "       [0.09049766, 0.9095023 ],\n",
              "       [0.45303786, 0.5469622 ],\n",
              "       [0.12543945, 0.87456053],\n",
              "       [0.94064754, 0.05935246],\n",
              "       [0.05008588, 0.9499141 ],\n",
              "       [0.6757894 , 0.32421055],\n",
              "       [0.18535908, 0.81464094],\n",
              "       [0.7509424 , 0.24905755],\n",
              "       [0.04295469, 0.95704526],\n",
              "       [0.7148519 , 0.28514808],\n",
              "       [0.24762939, 0.7523706 ],\n",
              "       [0.82708263, 0.1729173 ],\n",
              "       [0.19970109, 0.80029887],\n",
              "       [0.87371886, 0.12628116],\n",
              "       [0.28359118, 0.71640885],\n",
              "       [0.5450657 , 0.4549343 ],\n",
              "       [0.36411208, 0.6358879 ],\n",
              "       [0.94367784, 0.05632218],\n",
              "       [0.36411208, 0.6358879 ],\n",
              "       [0.942217  , 0.05778298],\n",
              "       [0.38573614, 0.6142639 ],\n",
              "       [0.942836  , 0.05716397],\n",
              "       [0.3430234 , 0.65697664],\n",
              "       [0.5450657 , 0.4549343 ],\n",
              "       [0.3430234 , 0.65697664],\n",
              "       [0.82708263, 0.1729173 ],\n",
              "       [0.10654986, 0.89345014],\n",
              "       [0.942217  , 0.05778298],\n",
              "       [0.4078206 , 0.5921795 ],\n",
              "       [0.87371886, 0.12628116],\n",
              "       [0.38573614, 0.6142639 ]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 60
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vfGswUXM4vFD",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "rounded_predictions = model.predict_classes(scaled_test_samples, batch_size=10, verbose=0)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OFjyjksp5FIb",
        "colab_type": "code",
        "outputId": "f8be6ed7-54d7-4218-dcac-0caecc6a31e3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 357
        }
      },
      "source": [
        "rounded_predictions"
      ],
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 1, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1, 0, 1,\n",
              "       0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1,\n",
              "       0, 1])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 59
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I0BJRxspxMlb",
        "colab_type": "text"
      },
      "source": [
        "### Confusion Matrix"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2cBGvG_Aw3IV",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "%matplotlib inline\n",
        "from sklearn.metrics import confusion_matrix\n",
        "import itertools\n",
        "import matplotlib.pyplot as plt\n",
        "\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BhO0Jey0x73y",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "cm = confusion_matrix(test_labels, rounded_predictions)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zNa0cXLg5Ffs",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def plot_confusion_matrix(cm, classes,\n",
        "                          normalize=False,\n",
        "                          title='Confusion Matrix',\n",
        "                          cmap=plt.cm.Blues):\n",
        "  \n",
        "  plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
        "  plt.title(title)\n",
        "  plt.colorbar()\n",
        "  tick_marks = np.arange(len(classes))\n",
        "  plt.xticks(tick_marks, classes, rotation=45)\n",
        "  plt.yticks(tick_marks, classes)\n",
        "\n",
        "  if normalize:\n",
        "    cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
        "    print('Normalized Confusion Matrix')\n",
        "  else:\n",
        "    print('Confusion Matrix, without normalization')\n",
        "\n",
        "\n",
        "  print(cm)\n",
        "\n",
        "  thresh = cm.max()/2.\n",
        "  for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
        "    plt.text(i, j, cm[i,j],\n",
        "             horizontalalignment='center',\n",
        "             color='white' if cm[i,j] > thresh else 'black')\n",
        "    \n",
        "  plt.tight_layout()\n",
        "  plt.ylabel('True label')\n",
        "  plt.xlabel('Predicted label')\n",
        "\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cSZkD8c-yHAZ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 362
        },
        "outputId": "8aea268e-6606-4e38-da86-28514f0dd7c2"
      },
      "source": [
        "cm_plot_labels = ['no_side_effects','had_side_effects']\n",
        "plot_confusion_matrix(cm, cm_plot_labels)"
      ],
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Confusion Matrix, without normalization\n",
            "[[190  20]\n",
            " [ 10 200]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAV0AAAEmCAYAAADBbUO1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CJgGIw6vCON5",
        "colab_type": "text"
      },
      "source": [
        "### Safe and Load Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Bg6RLpoGCSBd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model.save('medical_trial_model.h5')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kJpEnLZ_CzWJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras.models import load_model\n",
        "new_model = load_model('medical_trial_model.h5')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kolcPTvaDCpo",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 255
        },
        "outputId": "50cdd69c-38cd-45ef-bfd6-5db2a7ec052f"
      },
      "source": [
        "new_model.summary()"
      ],
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_3\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense_7 (Dense)              (None, 16)                32        \n",
            "_________________________________________________________________\n",
            "dense_8 (Dense)              (None, 32)                544       \n",
            "_________________________________________________________________\n",
            "dense_9 (Dense)              (None, 2)                 66        \n",
            "=================================================================\n",
            "Total params: 642\n",
            "Trainable params: 642\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "l5hKLwDyDLmd",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "70aa9660-9772-4953-e611-408fac0c78ec"
      },
      "source": [
        "new_model.get_weights()"
      ],
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[array([[ 0.6648388 ,  0.50189036,  0.38900763, -0.22821629, -0.21453881,\n",
              "         -0.10831296,  0.44520137, -0.0807988 , -0.5859293 ,  0.19232783,\n",
              "         -0.19289225, -0.4477049 , -0.45036012, -0.42379594, -0.43707004,\n",
              "          0.60803145]], dtype=float32),\n",
              " array([-0.11805252, -0.10020076,  0.17407382,  0.        ,  0.        ,\n",
              "         0.17324878, -0.09475537,  0.        ,  0.        ,  0.01189806,\n",
              "         0.        ,  0.        ,  0.        ,  0.        ,  0.        ,\n",
              "        -0.09736551], dtype=float32),\n",
              " array([[-3.01950902e-01, -6.70783445e-02,  5.04308641e-01,\n",
              "          2.45550528e-01,  3.37130487e-01,  3.49270999e-02,\n",
              "          1.48625225e-01, -4.21856672e-01, -2.11013898e-01,\n",
              "          1.41238004e-01, -1.42290249e-01, -1.25860602e-01,\n",
              "          4.87779558e-01, -3.20591003e-01,  8.14747810e-02,\n",
              "         -2.66698655e-02, -1.91592485e-01, -4.65623289e-01,\n",
              "         -2.67674297e-01, -3.47909570e-01, -1.57355577e-01,\n",
              "          3.88665020e-01, -4.75638270e-01, -1.66096538e-01,\n",
              "          4.15881038e-01,  4.06076014e-01,  1.38684034e-01,\n",
              "          5.35236381e-04, -8.57473835e-02,  1.75286591e-01,\n",
              "          4.52870429e-01,  1.44126445e-01],\n",
              "        [-3.20751488e-01, -2.43416473e-01,  4.44130629e-01,\n",
              "          4.94737595e-01,  7.08091706e-02,  2.40056783e-01,\n",
              "         -3.32004219e-01, -1.18009880e-01, -2.85578907e-01,\n",
              "         -2.93711603e-01, -1.93523213e-01,  2.30406791e-01,\n",
              "          3.94924104e-01,  1.60294980e-01, -3.22102666e-01,\n",
              "          3.15100819e-01, -5.92075586e-02, -4.69914734e-01,\n",
              "         -2.72195876e-01, -8.97802114e-02,  6.67130575e-03,\n",
              "         -2.91099828e-02,  4.71989959e-02,  3.23939323e-02,\n",
              "          1.55611411e-01,  5.03432572e-01,  1.72615319e-01,\n",
              "          1.15170948e-01,  1.82308495e-01, -4.54224348e-01,\n",
              "          2.05257744e-01,  1.77711710e-01],\n",
              "        [-3.29476833e-01,  1.75827146e-02,  4.26669329e-01,\n",
              "          1.94797486e-01,  1.07175685e-01, -3.53202730e-01,\n",
              "         -1.59528598e-01,  2.73117870e-01,  1.80750504e-01,\n",
              "          2.42397949e-01,  1.96337819e-01,  1.26543094e-03,\n",
              "          4.36029434e-01, -1.50420666e-01, -2.79505789e-01,\n",
              "          3.86168897e-01, -2.04067379e-01,  2.11576030e-01,\n",
              "          2.60212511e-01, -2.63218284e-01,  1.54744640e-01,\n",
              "         -1.57294929e-01,  1.13366246e-01,  5.65041490e-02,\n",
              "          1.10578261e-01,  1.56800672e-01, -1.19216591e-01,\n",
              "          9.06303003e-02,  1.97796345e-01,  2.51285166e-01,\n",
              "          1.58896178e-01,  6.37583993e-03],\n",
              "        [-1.75891861e-01,  1.40000015e-01, -3.53195816e-01,\n",
              "          2.54923671e-01,  3.57507169e-02, -1.70932531e-01,\n",
              "          3.21475774e-01, -8.19396377e-02,  3.67266834e-02,\n",
              "          2.35528558e-01, -1.07755408e-01, -1.60053834e-01,\n",
              "          1.88332409e-01,  3.52756053e-01,  1.17898315e-01,\n",
              "         -3.01986456e-01,  2.62375176e-02,  3.51090580e-01,\n",
              "          3.47586542e-01, -3.04887444e-01, -1.59262717e-02,\n",
              "          1.43535316e-01,  3.03489953e-01, -2.27655470e-02,\n",
              "         -1.46549478e-01,  1.81813985e-01, -1.83376133e-01,\n",
              "          2.99573809e-01, -2.14595824e-01,  4.34887409e-03,\n",
              "         -2.68950671e-01, -2.30259359e-01],\n",
              "        [ 1.27083957e-01, -1.10371798e-01, -3.27576518e-01,\n",
              "          2.30259210e-01, -3.33536148e-01, -2.35348254e-01,\n",
              "          2.99318999e-01, -2.06967488e-01, -3.30213904e-01,\n",
              "          6.71264529e-03, -1.44045204e-01, -2.13149503e-01,\n",
              "          2.04182118e-01, -2.76654363e-01,  1.51709110e-01,\n",
              "          3.24423581e-01,  5.58200479e-02,  3.13666135e-01,\n",
              "          2.82410830e-01,  2.27195948e-01,  2.03387111e-01,\n",
              "          4.72483039e-03,  3.20382714e-02,  1.63097233e-01,\n",
              "         -2.74015725e-01,  3.04377884e-01, -3.15460891e-01,\n",
              "         -3.14372003e-01, -2.02144548e-01,  2.16008216e-01,\n",
              "         -2.01112613e-01,  6.32256567e-02],\n",
              "        [-1.95933536e-01,  1.36937693e-01, -6.04170933e-02,\n",
              "         -2.11000159e-01,  1.83877155e-01, -1.46025345e-01,\n",
              "          1.18674636e-02,  1.45631237e-02,  3.55658114e-01,\n",
              "          4.66292888e-01,  3.39565903e-01,  1.59771115e-01,\n",
              "         -3.47162560e-02, -2.45579064e-01,  4.13451433e-01,\n",
              "         -5.06250225e-02,  2.72548884e-01,  4.89889383e-01,\n",
              "          3.17556947e-01, -2.73980230e-01,  1.41907111e-01,\n",
              "         -2.34876554e-02,  6.09383166e-01,  6.20325327e-01,\n",
              "          9.70536694e-02, -1.97530299e-01,  3.30912828e-01,\n",
              "          1.23553522e-01,  4.20023233e-01,  2.47292802e-01,\n",
              "         -2.24734768e-01,  1.04097947e-01],\n",
              "        [ 1.63163096e-01, -3.66103917e-01,  3.90915751e-01,\n",
              "          3.59424770e-01,  2.43808359e-01,  2.41462320e-01,\n",
              "          3.25977713e-01,  1.07448608e-01, -7.29668066e-02,\n",
              "          2.05816273e-02, -3.89296681e-01, -1.73842117e-01,\n",
              "          2.57745802e-01, -1.29212886e-01, -1.46867752e-01,\n",
              "          3.69756520e-02, -1.48208633e-01, -2.81738371e-01,\n",
              "         -4.60706383e-01,  2.42634445e-01, -3.02751213e-01,\n",
              "          4.08929408e-01, -3.89094390e-02, -2.34923258e-01,\n",
              "         -1.19824387e-01,  3.61584395e-01, -3.48476678e-01,\n",
              "          1.31356925e-01, -3.82225871e-01, -4.60338265e-01,\n",
              "          5.01464427e-01,  5.64092398e-02],\n",
              "        [-8.79713595e-02, -2.53936768e-01,  1.45729035e-01,\n",
              "         -3.22868794e-01, -2.69633949e-01, -2.33466730e-01,\n",
              "          2.12673575e-01, -1.35778859e-01, -2.69167811e-01,\n",
              "          2.79249161e-01,  3.20087761e-01, -2.70711899e-01,\n",
              "         -3.13257456e-01,  3.06950659e-01,  3.35623026e-02,\n",
              "          2.65479892e-01,  9.01130736e-02, -1.07035458e-01,\n",
              "         -4.39871550e-02, -7.02641904e-02,  2.70481080e-01,\n",
              "          8.95660222e-02,  2.11141139e-01,  2.88226634e-01,\n",
              "          2.95191407e-02,  2.27960736e-01, -2.78460354e-01,\n",
              "          3.78749371e-02, -6.50574565e-02, -1.14697590e-01,\n",
              "          2.54203826e-01, -1.16511256e-01],\n",
              "        [-6.74010813e-02, -1.36825606e-01,  5.01157939e-02,\n",
              "         -3.08658063e-01, -9.11500752e-02,  2.27625757e-01,\n",
              "          8.03666115e-02,  2.85535604e-01, -2.78415591e-01,\n",
              "          2.44776219e-01,  1.51085407e-01, -1.18059218e-01,\n",
              "         -4.80289459e-02,  1.48936778e-01, -2.22241253e-01,\n",
              "          1.13625884e-01, -2.22262159e-01,  2.03071833e-02,\n",
              "         -1.20859712e-01,  3.45474154e-01,  2.53971666e-01,\n",
              "         -3.24067473e-03, -1.63701221e-01,  3.15259010e-01,\n",
              "          9.91087258e-02,  1.63823038e-01, -2.23278314e-01,\n",
              "         -1.04674146e-01, -2.15718195e-01,  2.42180198e-01,\n",
              "          1.88072622e-02,  1.70815378e-01],\n",
              "        [ 1.54663533e-01,  4.99781556e-02, -1.88546434e-01,\n",
              "         -1.41767124e-02, -2.75558919e-01, -1.48378909e-02,\n",
              "         -3.22415531e-01, -1.78822771e-01,  9.27607045e-02,\n",
              "          1.31649897e-01,  1.69861093e-01,  1.84392542e-01,\n",
              "         -1.54301718e-01,  5.31028509e-02,  1.97345242e-01,\n",
              "          1.60085604e-01,  7.05743358e-02,  1.50762815e-02,\n",
              "          7.65730217e-02, -2.25418806e-03,  2.50152349e-01,\n",
              "          2.54655868e-01, -1.19049795e-01, -2.29278475e-01,\n",
              "         -5.99163724e-03,  3.34818840e-01, -4.27809566e-01,\n",
              "         -1.73059046e-01, -1.14590123e-01,  5.87705597e-02,\n",
              "          1.70636430e-01, -2.94604331e-01],\n",
              "        [ 2.36829817e-02,  8.95210803e-02, -3.47479105e-01,\n",
              "         -3.07961613e-01, -3.04433256e-01,  3.37903589e-01,\n",
              "         -2.68313587e-01, -3.86597812e-02,  2.12447673e-01,\n",
              "         -3.43159467e-01, -2.14672446e-01, -1.74392030e-01,\n",
              "          1.34158969e-01,  6.24655187e-02,  2.73954421e-01,\n",
              "         -4.07520533e-02, -2.42405832e-01,  5.65753281e-02,\n",
              "          2.29393393e-01,  1.49966091e-01, -2.59619683e-01,\n",
              "          4.22764122e-02,  3.17892998e-01,  9.13515985e-02,\n",
              "         -5.08865118e-02,  1.16573215e-01,  3.09680700e-02,\n",
              "         -6.65508807e-02,  4.83399630e-03, -3.25021982e-01,\n",
              "          2.75776118e-01, -1.26594886e-01],\n",
              "        [-2.74275839e-01, -2.47198254e-01,  2.77680963e-01,\n",
              "          1.35453910e-01,  2.00914592e-01, -1.60744026e-01,\n",
              "          2.00198144e-01, -2.07402274e-01,  1.82753116e-01,\n",
              "          2.70476788e-01, -2.90890276e-01, -3.06011915e-01,\n",
              "          9.95618105e-03,  1.43543571e-01,  2.10092098e-01,\n",
              "          3.39795023e-01, -1.96833313e-02,  4.63794172e-02,\n",
              "          3.24015170e-01,  1.89933687e-01, -1.21218458e-01,\n",
              "         -1.64820299e-01,  3.53186041e-01,  2.41611004e-02,\n",
              "         -2.36212432e-01,  2.62347251e-01,  1.17524892e-01,\n",
              "         -2.06160709e-01,  2.09866613e-01, -1.04908645e-01,\n",
              "          2.57770807e-01, -1.23136237e-01],\n",
              "        [-2.16351986e-01,  9.33274627e-02, -2.83978224e-01,\n",
              "         -3.16770911e-01,  2.81898379e-02,  2.99953967e-01,\n",
              "          1.69520289e-01,  2.68143922e-01,  4.65838015e-02,\n",
              "          3.22750509e-02,  8.39901567e-02,  2.20502049e-01,\n",
              "         -2.79603362e-01, -3.52526605e-01, -1.89484388e-01,\n",
              "         -1.05960205e-01, -1.45512581e-01,  2.94773310e-01,\n",
              "          1.87054604e-01, -2.61680484e-02, -3.53231370e-01,\n",
              "         -2.84339160e-01,  3.53161484e-01, -3.02108526e-02,\n",
              "          2.01002389e-01, -6.79565668e-02,  2.04882890e-01,\n",
              "          7.86461830e-02,  8.88660252e-02, -3.16205114e-01,\n",
              "         -2.55880177e-01, -2.54257619e-01],\n",
              "        [-1.71535477e-01,  1.69911683e-02,  7.33585358e-02,\n",
              "          2.95963734e-01, -3.19700778e-01, -3.33052069e-01,\n",
              "          3.29243332e-01,  3.15293998e-01,  2.44136781e-01,\n",
              "         -3.84847820e-02,  5.29831350e-02,  1.64683670e-01,\n",
              "         -4.21971679e-02,  1.83786899e-01, -1.46080717e-01,\n",
              "          2.26369649e-01,  5.05205095e-02, -5.71817160e-02,\n",
              "          1.32162482e-01, -3.51794422e-01, -1.25062600e-01,\n",
              "         -3.09983909e-01, -1.37552142e-01, -1.83366433e-01,\n",
              "         -1.30655989e-01,  2.12196559e-01, -2.95128912e-01,\n",
              "         -3.28611225e-01,  2.67843872e-01,  1.57875150e-01,\n",
              "          2.09898144e-01,  2.55072802e-01],\n",
              "        [ 2.25859493e-01, -3.04787964e-01,  2.81010479e-01,\n",
              "          3.02786440e-01, -3.03355396e-01,  3.49512488e-01,\n",
              "          1.75643116e-01,  2.87433833e-01,  2.99597651e-01,\n",
              "          2.59391993e-01, -2.07567319e-01, -3.50073755e-01,\n",
              "          2.49713510e-01,  2.80038387e-01, -2.93752074e-01,\n",
              "         -1.39200673e-01,  2.24227875e-01,  8.95765424e-02,\n",
              "          2.73110658e-01, -2.29758978e-01, -1.11551076e-01,\n",
              "          8.27704966e-02,  3.35804909e-01,  2.01760441e-01,\n",
              "         -2.08130330e-01, -1.92959905e-01,  3.20984155e-01,\n",
              "          1.68158025e-01,  1.72094375e-01,  3.10258180e-01,\n",
              "         -5.24905324e-02,  2.98968047e-01],\n",
              "        [-3.84046435e-02,  5.13363183e-02,  3.81663561e-01,\n",
              "         -2.62885485e-02,  1.32509470e-01, -2.98067063e-01,\n",
              "          1.08109623e-01, -2.04048380e-01, -1.89730838e-01,\n",
              "          9.24592316e-02, -3.88286375e-02, -2.31012270e-01,\n",
              "          4.39201027e-01, -1.10202536e-01,  2.16268092e-01,\n",
              "          2.69317627e-01, -2.81551093e-01,  2.21702144e-01,\n",
              "          1.81210861e-01, -1.22294396e-01,  4.17545252e-02,\n",
              "          3.42278838e-01, -7.03699961e-02, -2.21774071e-01,\n",
              "          2.78334051e-01,  3.64449590e-01,  1.49867803e-01,\n",
              "         -8.98181871e-02, -1.70156941e-01, -3.17600310e-01,\n",
              "         -1.55838400e-01, -2.57384598e-01]], dtype=float32),\n",
              " array([ 0.        ,  0.15754682, -0.09404787, -0.06187629, -0.00391461,\n",
              "         0.        ,  0.        ,  0.18150027,  0.16361175,  0.09086663,\n",
              "         0.11117079, -0.03316224, -0.08668511,  0.        ,  0.09480175,\n",
              "        -0.09270778, -0.01273907,  0.21783397,  0.11157828,  0.        ,\n",
              "         0.09330042, -0.04746203,  0.16243994,  0.19147992, -0.07180237,\n",
              "        -0.07150049,  0.08499099,  0.07506641,  0.09944201,  0.19061355,\n",
              "        -0.05832192, -0.01453693], dtype=float32),\n",
              " array([[ 0.40035334,  0.15715572],\n",
              "        [ 0.23669066, -0.38226682],\n",
              "        [-0.08523382,  0.29905516],\n",
              "        [-0.08310172,  0.52679455],\n",
              "        [ 0.1600025 ,  0.24052101],\n",
              "        [ 0.3191693 , -0.07810754],\n",
              "        [-0.05744097, -0.14579496],\n",
              "        [ 0.53229874, -0.5129938 ],\n",
              "        [ 0.38871297, -0.76132065],\n",
              "        [ 0.36036736, -0.25090757],\n",
              "        [ 0.5815614 , -0.29242042],\n",
              "        [ 0.14683227,  0.26670358],\n",
              "        [-0.40538046,  0.5728676 ],\n",
              "        [ 0.00190255, -0.2143347 ],\n",
              "        [ 0.44804105, -0.2054837 ],\n",
              "        [-0.27093714,  0.30144444],\n",
              "        [ 0.13115582,  0.35450628],\n",
              "        [ 0.35767716, -0.34538233],\n",
              "        [ 0.6087771 , -0.5195047 ],\n",
              "        [-0.16043875,  0.01065779],\n",
              "        [ 0.13732795, -0.36528078],\n",
              "        [-0.28466386,  0.2537653 ],\n",
              "        [ 0.62538224, -0.42798197],\n",
              "        [ 0.6204732 , -0.58860993],\n",
              "        [-0.39519504,  0.5076901 ],\n",
              "        [-0.5655293 ,  0.45011586],\n",
              "        [ 0.15586688, -0.08351792],\n",
              "        [ 0.05742649, -0.04503738],\n",
              "        [ 0.13661668, -0.25501832],\n",
              "        [ 0.5138663 , -0.2831514 ],\n",
              "        [-0.21764491,  0.60757893],\n",
              "        [-0.2287703 , -0.22743273]], dtype=float32),\n",
              " array([ 0.09396163, -0.09396163], dtype=float32)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2LcJw6z2DQ7a",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "4662e79b-619d-45e3-a38b-5d831685a245"
      },
      "source": [
        "new_model.optimizer"
      ],
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.optimizers.Adam at 0x7f54fc6ba438>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 68
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3O4cv6QSDbT9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}