Deep_Learning_with_TensorFlow.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Deep Learning with TensorFlow",
      "provenance": [],
      "authorship_tag": "ABX9TyNpfNH37lmBzvjedEiTuflP"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "5YFhCappeZtx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import tensorflow as tf"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hh2_cMZyeccR",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from tensorflow.keras.datasets import mnist\n",
        "from collections import Counter\n",
        "\n",
        "import sklearn as sk\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "%matplotlib inline"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6xZyJb3ze0D3",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mWnqKEywe-jg",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "8f400c3e-f260-4c00-c5c8-f07a0a4e29d8"
      },
      "source": [
        "# Get tensorflow version\n",
        "print(tf.__version__)"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "1.15.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Zc5d6HNFfFeZ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "f64cab68-83e2-42ce-a827-adbb666c336b"
      },
      "source": [
        "# Create regression dataset: parabolic curve\n",
        "X = np.linspace(-3, 3, 100).reshape(-1,1)\n",
        "y = X[:,0] ** 2\n",
        "print(X.shape)\n",
        "print(y.shape)"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(100, 1)\n",
            "(100,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zXFr3T28f2VV",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "outputId": "468981c9-b897-404e-aebb-66fca4a8a1c6"
      },
      "source": [
        "# Plot curve\n",
        "plt.scatter(X,y)\n",
        "plt.show()"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWoAAAD4CAYAAADFAawfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjMsIGh0\ndHA6Ly9tYXRwbG90bGliLm9yZy+AADFEAAATyElEQVR4nO3db4xc1XnH8d/j9ZKsScoisS/ixa79\nAtGggHC1olFdRSkkMaIodlCjkjaVmqiyojYtoITUNFJQW0W4cpUm6ouqFtA/Ciqg4FpVTWtamaoF\nFcoaOyVgHCFa/qzTxmlZJeBtWcPTFzPX3F3P7NyZuX/OOff7kSzt7Myuz7Xn/Pbsc557r7m7AADh\nWtf0AAAAayOoASBwBDUABI6gBoDAEdQAELj1VXzTSy65xLds2VLFtwaAJB09evQH7j7T67lKgnrL\nli2an5+v4lsDQJLM7KV+z1H6AIDAEdQAEDiCGgACR1ADQOAIagAIHEENAIGrpD1vFAePLWjf4ZM6\ntbikjdNTun3H5dq1bbbpYQHAQFXnVxBBffDYgu448IyWlt+SJC0sLumOA89IEmENIGh15FcQpY99\nh0+eO8jM0vJb2nf4ZEMjAoBi6sivIIL61OLSUJ8HgFDUkV9BBPXG6amhPg8Aoagjv4II6tt3XK6p\nyYkVn5uanNDtOy5vaEQAUEwd+RXEZmJWcKfrA0Bs6sgvq+LmtnNzcz7O1fNo1QMQurJzysyOuvtc\nr+eCWFHn0aoHIHR151QQNeo8WvUAhK7unAouqGnVAxC6unMquKCmVQ9A6OrOqeCCmlY9AKGrO6eC\n20ykVQ9A6OrOqSDb8wCgbaJqz1uNnmoAoWgqj4IOanqqAYSiyTwKbjMxj55qAKFoMo+CDmp6qgGE\nosk8Cjqo6akGEIom8yjooKanGkAomsyjQpuJZnabpF+V5JKekfQZd//fKgcm0VMNIBxN5tHAPmoz\nm5X0mKQr3H3JzB6U9LC7/1m/r6GPGgCGU0Yf9XpJU2a2LGmDpFNlDW4Y9FQDqFsIuTMwqN19wcz+\nQNLLkpYkPeLuj6x+nZntlrRbkjZv3lz2OOmpBlC7UHJn4GaimV0saaekrZI2SrrQzD69+nXuvt/d\n59x9bmZmpvSB0lMNoG6h5E6Rro+PSPp3dz/t7suSDkj66WqHdT56qgHULZTcKRLUL0v6oJltMDOT\ndJ2kE9UO63z0VAOoWyi5MzCo3f1JSd+S9LQ6rXnrJO2veFznoacaQN1CyZ1CXR/ufqekOysey5ro\nqQZQt1ByJ+ir5622a9vsin+gg8cWtH3vEYIbQGl6teM9vufaRscUVVDnhdI2AyAdoeZK0Nf6WEso\nbTMA0hFqrkQb1KG0zQBIR6i5Em1Qh9I2AyAdoeZKtEEdStsMgHSEmivRbiaG0jYDIB2h5srAy5yO\noonLnIZwhSsAcQohP8q4zGnQQm2pARC+GPIj2hp1XqgtNQDCF0N+JBHUobbUAAhfDPmRRFCH2lID\nIHwx5EcSQR1qSw2A8MWQH0lsJobaUgMgfDHkRzLteXkhtNoACFtoOZF8e15eDK02AJoVW04kUaPO\ni6HVBkCzYsuJ5II6hlYbAM2KLSeSC+oYWm0ANCu2nEguqGNotQHQrNhyIrnNxBhabQA0K7acSLI9\nLy+0FhwAzQk5D1rVnpcXWwsOgOrEnAfJ1ajzYmvBAVCdmPMg6aCOrQUHQHVizoOkgzq2FhwA1Yk5\nD5IO6thacABUJ+Y8SHozMbYWHADViTkPkm/Pywu5NQdANWKZ961tz8uLuTUHwGhSmfdJ16jzYm7N\nATCaVOZ9a4I65tYcAKNJZd63Jqhjbs0BMJpU5n1rgjrm1hwAo0ll3rdmMzHm1hwAo0ll3hdqzzOz\naUl3S/qAJJf0WXf/l36vD7U9Ly+Wlh0Aw4txfpfRnvcNSX/n7j9vZhdI2lDa6BqQSssOgPOlOL8H\n1qjN7CJJH5J0jyS5+5vuvlj1wKqUSssOgPOlOL+LbCZulXRa0p+a2TEzu9vMLlz9IjPbbWbzZjZ/\n+vTp0gdaplRadgCcL8X5XSSo10v6SUl/7O7bJL0hac/qF7n7fnefc/e5mZmZkodZrlRadgCcL8X5\nXSSoX5X0qrs/2X38LXWCO1qptOwAOF+K83tgULv7f0p6xcyyo7xO0nOVjqpiu7bN6q6brtTs9JRM\n0vTUpN49uU63PXBc2/ce0cFjC00PEcCQDh5b0Pa9R3TbA8f1rvXrdPGGSZmk2ekp3XXTldFuJErF\nuz5+Q9J93Y6PFyV9proh1WPXtlnt2jab5A4x0Dar5/Hi0rKmJif0h79wdRLzuNCZie5+vFt/vsrd\nd7n7a1UPrC4p7hADbZP6PG7NKeT9pLhDDLRN6vO49UGd4g4x0Dapz+PWB3WKO8RA26Q+j1tzUaZ+\nVl+05aKpSZlJtz1wXPsOn4ziGgFAW+Wv6XFRt3tr8cxyNNf3KKr1QS3RAQLEKPVOj7zWlz7yUt85\nBlLSpvlKUOekvnMMpKRN85Wgzkl95xhISZvmK0Gdk/rOMZCSNs1XNhNzUrltD9AGbZqvhW7FNawY\nbsVVRIy38wFSl+q8LONWXK1Dqx4QnrbOS2rUfbSp9QeIRVvnJUHdR5taf4BYtHVeEtR9tKn1B4hF\nW+clQd1Hr9YfU6cmxl1ggHpld29ZWFySrXou1Za8PDYT+8i3/mRvjqw/pi0bGEAIVm8gunRuPs4m\n1PWxFlbUa9i1bVaP77lWs9NTWt3E2IYNDCAEvTYQs5B+fM+1yYe0RFAX0tYNDCAEzD+CupC2bmAA\nIWD+EdSFtOmaAkBomH9sJhbCXWCA+rXl7i1FENQFcRcYoD5tuntLEZQ+htTWU1iBOjHPViKoh8QO\nNFA95tlKBPWQ2IEGqsc8W4mgHhI70ED1mGcrsZk4JDpAgOrQ6dEbQT0COkCA8tHp0R+ljzGwMw2U\nh/nUH0E9BnamgfIwn/ojqMfAzjRQHuZTfwT1GLi5ADC+tt8UoAg2E8fAzQWA8XBTgGJYUY+JmwsA\no+OmAMUUDmozmzCzY2b2N1UOKFZshADDY94UM8yK+hZJJ6oaSOzYCAGGx7wpplBQm9mlkn5O0t3V\nDidenPIKDI95U0zRzcSvS/qSpPf2e4GZ7Za0W5I2b948/sgiw6nlQHGcKj4cc1+9BbbqBWY3SrrB\n3X/NzD4s6YvufuNaXzM3N+fz8/PljTIyq3eypc4q4a6bruQNiNZjfvRmZkfdfa7Xc0VKH9slfdzM\n/kPS/ZKuNbNvlji+5HAqLNAf82N4A4Pa3e9w90vdfYukmyUdcfdPVz6yiLGTDfTH/BgefdQVYCcb\n6I/5Mbyhgtrd/3FQfRqcWg70wqnio+MU8gpwajmwEqeKj4fSR0U4tRx4B6eKj4egrhgbJwDzYFwE\ndcXYOAGYB+MiqCvGxiLaKts83LrnkN74v7OanFi5hcgGYnFsJlaMjUW0Ua8b1U6uM128YZJTxUdA\nUNcgu2t51pqUl20s8oZFSnptHi6/7dpwwXod+8rHGhpVvCh91IgNFbQF7/VyEdQ1YkMFbcF7vVwE\ndY16bSxOrjOdefOstu45xOYiosfZh9WgRl2jXtesfuPNs3rtzLIkNhcRN84+rA5BXbNsY1GStu89\nosWl5RXPs7mIWA06+xCjo/TRIDZckBLez9UhqBvUb2PFJerViEZWl+53ryg2EMdHUDeo1+ZiJqtX\nE9YIWVaXXn1+QIYNxHIQ1A3atW1Wd910pWb7rDi4yh5C16sunZmdnmr9fRDLwmZiw7LNxa17DvX8\n1ZH6HkLW7/1pEhuIJWJFHQhOEECMeN/Wg6AOBFfZQ0w4saVelD4CwVX2EAtObKkfK+qAcPsuxIDb\natWPoA4QJw4gZLw/60dQB4gTYRAiTmxpDkEdIE6EQWg4saVZBHWAOBEGoeHElmbR9REoToRBSDix\npVmsqANHvRpNoi4dBoI6cNSr0RTq0uEgqANHvRpNoS4dDmrUEaBejSZQlw4HK+qIUK9GHahLh4eg\njgj1alSNunSYCOqIUK9G1ahLh4kadWSoV6NK1KXDxIo6UtSrUSbq0mEbGNRmtsnMHjWz58zsWTO7\npY6BYW3Uq1EW6tLhK7KiPivpC+5+haQPSvp1M7ui2mFhEOrVKAt16fANDGp3/567P939+EeSTkji\nfy0A2Y0GVt8KKUO9GkUMqksT0s0bqkZtZlskbZP0ZI/ndpvZvJnNnz59upzRoRDq1RgFdel4FA5q\nM3uPpIck3eruP1z9vLvvd/c5d5+bmZkpc4wYgHo1hkVdOi6FgtrMJtUJ6fvc/UC1Q8KwqFdjWNSl\n41Kk68Mk3SPphLt/rfohYRSD6tULi0uUQXCu3NFvJU1dOkxFVtTbJf2ypGvN7Hj3zw0VjwsjWquu\nSBmk3QaVOyTq0qEq0vXxmLubu1/l7ld3/zxcx+AwvLXq1RJlkDZbq9whUZcOGaeQJyb7lXXf4ZN9\nV0607bXTWv/vs9NTun3H5ZQ8AsUp5AnK6tX9Nhdp22uXQW14s9NT1KUDR1AnjLY90IaXBoI6YbTt\ngTa8NBDUiaNtr51ow0sLQd0StO21B2146SGoW4K2vfagDS89BHVLDKpXS5RBYjeo3CFRl44VfdQt\nkt3Ga63JnJVBstcjDlm5Y62VdNaGh/iwom4hyiDpodyRNoK6hSiDpINyRztQ+mgpyiDxo9zRHqyo\nW44ySLwod7QHQd1ylEHiQ7mjfSh9gDJIRCh3tBMrapxTpAxy6wPHWV3XLFtBb91zSF948NuUO1qI\nFTXOKXIta4nVdZ1Wr6Df8n4XK+Wa0iljRY0VBl3LOsMmYz0GbRhmuKZ02ghq9DSoDCJ1VtZb9xyi\nFFKBIhuGGcod6aP0gZ6KlkFclELKVmTDcMJMb7trI+WOVjBfo+Y1qrm5OZ+fny/9+6IZRYIjQ510\ndAePLQz8wSh1VtC03qXHzI66+1yv51hRY6D86vrU4lLfe+9JrK5HVfSHIT8I24kVNYZWtHZKqAxW\ndBUt0R+durVW1GwmYmhFNhol7hwzSJE7sWTYMGw3VtQYyTArQYnVdR7/duiFGjVKl512XrS2Su26\nY5iNWTYNkWFFjbGxQhyMfyMMstaKmqBGaYZZLZo6PdgpB1I+nLPjHYRVdHtR+kAtip4kI70TWqmW\nRFb/0CoS0in/0MJ4WFGjEsOsrjMpBNWwJQ6JVTQ6KH2gEaOEVowlkVFKHJmYjhPVIqjRqFFW11LY\noT1OOEusonE+ghqNGzfYQgjtFI4B4SKoEZRRSiJ5dQbeuOGcIZwxyNhBbWbXS/qGpAlJd7v73rVe\nT1CjiFFLInlZeE5PTcpMWjyzrItyH691GdAshE8tLq34muzj184sjxXOEiUOFDdWUJvZhKTvSvqo\npFclPSXpU+7+XL+vIahRVFkr1rXkV+A/+xMzevT507X9fayiUdS4fdTXSHrB3V/sfrP7Je2U1Deo\ngaKyU9Gl6kI737P9zSdePu/zZSCcUaUiQT0r6ZXc41cl/VQ1w0Gb1RHaZSKcUZfSzkw0s92SdkvS\n5s2by/q2aKlQQ5twRhOKBPWCpE25x5d2P7eCu++XtF/q1KhLGR2g5kObcEbTigT1U5IuM7Ot6gT0\nzZJ+sdJRAX30Cu1xujbG6RoB6jIwqN39rJl9XtJhddrz7nX3ZysfGTBAPrT7yYf5xlzXR/aYEEYM\nOOEFAALAPRMBIGIENQAEjqAGgMAR1AAQOIIaAAJXSdeHmZ2W9NKIX36JpB+UOJwmpXIsqRyHxLGE\nKJXjkMY7lh9395leT1QS1OMws/l+LSqxSeVYUjkOiWMJUSrHIVV3LJQ+ACBwBDUABC7EoN7f9ABK\nlMqxpHIcEscSolSOQ6roWIKrUQMAVgpxRQ0AyCGoASBwQQa1mf2emf2bmR03s0fMbGPTYxqVme0z\ns+e7x/NXZjbd9JhGYWafNLNnzextM4uulcrMrjezk2b2gpntaXo8ozKze83s+2b2nabHMi4z22Rm\nj5rZc9331i1Nj2kUZvZuM/tXM/t29zh+p/S/I8QatZn9mLv/sPvxb0q6wt0/1/CwRmJmH5N0pHtd\n79+XJHf/rYaHNTQze7+ktyX9iaQvuns017E1swlJ35X0UXXu+fmUpE+5e3Q3aDazD0l6XdJfuPsH\nmh7POMzsfZLe5+5Pm9l7JR2VtCu2/xczM0kXuvvrZjYp6TFJt7j7E2X9HUGuqLOQ7rpQYd7btBB3\nf8Tdz3YfPqHOrcyi4+4n3P1k0+MY0TWSXnD3F939TUn3S9rZ8JhG4u7/JOl/mh5HGdz9e+7+dPfj\nH0k6oc7NtKPiHa93H052/5SaWUEGtSSZ2VfN7BVJvyTpK02PpySflfS3TQ+ihWYlvZJ7/KoiDISU\nmdkWSdskPdnsSEZjZhNmdlzS9yX9vbuXehyNBbWZ/YOZfafHn52S5O5fdvdNku6T9PmmxlnEoGPp\nvubLks6qczxBKnIcQNnM7D2SHpJ066rfpqPh7m+5+9Xq/MZ8jZmVWpYqcnPbSrj7Rwq+9D5JD0u6\ns8LhjGXQsZjZr0i6UdJ1HuKmQNcQ/yexWZC0Kff40u7n0LBuTfchSfe5+4GmxzMud180s0clXS+p\ntA3fIEsfZnZZ7uFOSc83NZZxmdn1kr4k6ePufqbp8bTUU5IuM7OtZnaBpJsl/XXDY2q97ibcPZJO\nuPvXmh7PqMxsJuvmMrMpdTatS82sULs+HpJ0uTpdBi9J+py7R7kCMrMXJL1L0n93P/VEjB0sZvYJ\nSX8kaUbSoqTj7r6j2VEVZ2Y3SPq6pAlJ97r7Vxse0kjM7C8lfVidy2n+l6Q73f2eRgc1IjP7GUn/\nLOkZdea6JP22uz/c3KiGZ2ZXSfpzdd5b6yQ96O6/W+rfEWJQAwDeEWTpAwDwDoIaAAJHUANA4Ahq\nAAgcQQ0AgSOoASBwBDUABO7/ARu89hlnwyMCAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uFj-xLKTf9rD",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Split, train and test\n",
        "from sklearn.model_selection import train_test_split"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bJlWeSjZgDXA",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_train, x_test, y_train, y_test = train_test_split(X,y, random_state=2020)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YgcTSc2IgJw_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Create model with 1 hidden layer, 128 neurons\n",
        "model = Sequential()\n",
        "model.add(Dense(128, activation='relu'))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9OilPDc2gYjX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Add final hidden layer with output prediction\n",
        "model.add(Dense(1))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UFaBEOSkgeUg",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Compile model before training -- add optimizer, loss function, metrics\n",
        "model.compile(\n",
        "    optimizer='adam',\n",
        "    loss='mean_squared_error',\n",
        "    metrics=['mse']\n",
        ")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xGzHnfEzgr4f",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "363dd50a-88e8-498b-9d0d-cbd6a73aafd2"
      },
      "source": [
        "% time\n",
        "# Start training\n",
        "history = model.fit(\n",
        "    x=y_train,\n",
        "    y=y_train,\n",
        "    verbose=0,\n",
        "    validation_data=(x_test, y_test),\n",
        "    batch_size=100,\n",
        "    epochs=10000\n",
        ")"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "CPU times: user 4 µs, sys: 0 ns, total: 4 µs\n",
            "Wall time: 7.63 µs\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Jg9QakMkhAPb",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 225
        },
        "outputId": "e6deaa6e-8400-4b81-c9dc-726b7837adff"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense (Dense)                multiple                  256       \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              multiple                  129       \n",
            "=================================================================\n",
            "Total params: 385\n",
            "Trainable params: 385\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "D2Ybimj_hGqB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Save model layers into figure\n",
        "tf.keras.utils.plot_model(model, 'regression.png')\n",
        "ypred = model.predict(X)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cpe7MBMYhVD-",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "outputId": "1a068f28-7a3a-4d09-a889-50ef61cebd58"
      },
      "source": [
        "plt.plot(X, ypred,'r')\n",
        "plt.scatter(X,y)\n",
        "plt.show()"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWoAAAD4CAYAAADFAawfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uDrNQwghhbWk",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Linear Regression\n",
        "from sklearn import datasets\n",
        "x2, y2 = datasets.make_regression(n_samples=100,\n",
        "                                  n_features=1,\n",
        "                                  n_informative=1,\n",
        "                                  noise=1.0,\n",
        "                                  random_state=2020\n",
        "                                  )"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R6F4-u_Eiz1R",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "outputId": "1f15091a-50ab-4299-d7f1-a3ea02b08aac"
      },
      "source": [
        "plt.scatter(x2[:,0], y2)"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.collections.PathCollection at 0x7fc5a3333cf8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAD4CAYAAAAJmJb0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nHgAy7P8BPhbbf1s/9hMAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o7kWzC6Djqrr",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6lh39jmNnZ1g",
        "colab_type": "text"
      },
      "source": [
        "**Convolution Neural Network**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NuBGc2hJndjK",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "outputId": "c4b5fee9-d24f-4ab0-901d-39cec9a05b22"
      },
      "source": [
        "# Image Classification\n",
        "# Load data\n",
        "from tensorflow.keras.datasets import mnist\n",
        "(x_train, y_train), (x_test, y_test) = mnist.load_data()"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "11493376/11490434 [==============================] - 0s 0us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tdGSptBGnrh4",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "214073bf-d6d9-4be5-bd54-a9cff82769e4"
      },
      "source": [
        "print(x_train.shape)"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(60000, 28, 28)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VweBRPoInusr",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "40a8cc53-6198-4651-d5f9-528703a6167a"
      },
      "source": [
        "print(y_train.shape)"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(60000,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NW4BPUKWnvyb",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "outputId": "6bc29466-2a30-4ac6-b48b-8b793ce8be4e"
      },
      "source": [
        "# Show image\n",
        "plt.imshow(x_train[0], cmap='gray')"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7fc5a2e77550>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEARhB4Ig7EAQ/w8ie3GmjcGk5QAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0Mv0XaNXn3tL",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import collections"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Lf55qyj-oKck",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 190
        },
        "outputId": "327b3cfd-5601-4c45-906e-f141818bdd31"
      },
      "source": [
        "# Distributon of the labels in our training dataset\n",
        "label_counts = collections.Counter(y_train)\n",
        "label_counts"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Counter({0: 5923,\n",
              "         1: 6742,\n",
              "         2: 5958,\n",
              "         3: 6131,\n",
              "         4: 5842,\n",
              "         5: 5421,\n",
              "         6: 5918,\n",
              "         7: 6265,\n",
              "         8: 5851,\n",
              "         9: 5949})"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LP5NMlfmoSGa",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Image Dimensions --4:\n",
        "# [batch_size, image_height, image_weight, image_depth]"
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}