TF2_0_Linear_Regression.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyOsqNZjTAgTliayfhQJZYLP",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/sreya-rc/TF2.0/blob/main/TF2_0_Linear_Regression.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Part 1: Creating Model**\n",
        "This goes with lecture \"Regression Notebook\""
      ],
      "metadata": {
        "id": "xcZuNB8_b4gE"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "lYfv6mv1hXtB"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Other imports\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt"
      ],
      "metadata": {
        "id": "ZlZksjswiZ6z"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Get the data\n",
        "!wget https://raw.githubusercontent.com/lazyprogrammer/machine_learning_examples/master/tf2.0/moore.csv"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AKosBkzuikkp",
        "outputId": "3db4a9e3-4b36-48d4-ee28-3dafdeb51ff6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2023-07-29 20:59:18--  https://raw.githubusercontent.com/lazyprogrammer/machine_learning_examples/master/tf2.0/moore.csv\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 2302 (2.2K) [text/plain]\n",
            "Saving to: ‘moore.csv.2’\n",
            "\n",
            "\rmoore.csv.2           0%[                    ]       0  --.-KB/s               \rmoore.csv.2         100%[===================>]   2.25K  --.-KB/s    in 0s      \n",
            "\n",
            "2023-07-29 20:59:18 (26.1 MB/s) - ‘moore.csv.2’ saved [2302/2302]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!head moore.csv"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sj0sTeePjbXc",
        "outputId": "2ef380cf-5b9b-4c56-ebef-ef3e20b9676d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1971,2300\n",
            "1972,3500\n",
            "1973,2500\n",
            "1973,2500\n",
            "1974,4100\n",
            "1974,4500\n",
            "1974,8000\n",
            "1975,3510\n",
            "1976,5000\n",
            "1976,8500\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Load in the data\n",
        "data = pd.read_csv('moore.csv', header=None).to_numpy()"
      ],
      "metadata": {
        "id": "1gLiPOYIjCvX"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2D7HdrNDjvS5",
        "outputId": "ea85fbab-7e84-49e3-e032-0cb9de147569"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[       1971,        2300],\n",
              "       [       1972,        3500],\n",
              "       [       1973,        2500],\n",
              "       [       1973,        2500],\n",
              "       [       1974,        4100],\n",
              "       [       1974,        4500],\n",
              "       [       1974,        8000],\n",
              "       [       1975,        3510],\n",
              "       [       1976,        5000],\n",
              "       [       1976,        8500],\n",
              "       [       1976,        6500],\n",
              "       [       1978,        9000],\n",
              "       [       1978,       29000],\n",
              "       [       1979,       17500],\n",
              "       [       1979,       29000],\n",
              "       [       1979,       68000],\n",
              "       [       1981,       11500],\n",
              "       [       1982,       55000],\n",
              "       [       1982,      134000],\n",
              "       [       1983,       22000],\n",
              "       [       1984,       63000],\n",
              "       [       1984,      190000],\n",
              "       [       1985,      275000],\n",
              "       [       1985,       25000],\n",
              "       [       1985,       16000],\n",
              "       [       1986,      110000],\n",
              "       [       1986,      375000],\n",
              "       [       1986,       30000],\n",
              "       [       1987,      385000],\n",
              "       [       1987,      730000],\n",
              "       [       1987,      273000],\n",
              "       [       1987,      553000],\n",
              "       [       1988,      180000],\n",
              "       [       1988,      250000],\n",
              "       [       1989,      600000],\n",
              "       [       1989,     1000000],\n",
              "       [       1989,     1180235],\n",
              "       [       1989,      310000],\n",
              "       [       1990,     1200000],\n",
              "       [       1991,     1350000],\n",
              "       [       1991,       35000],\n",
              "       [       1992,      600000],\n",
              "       [       1992,      900000],\n",
              "       [       1993,     2800000],\n",
              "       [       1993,     3100000],\n",
              "       [       1994,      578977],\n",
              "       [       1994,     2500000],\n",
              "       [       1995,     2500000],\n",
              "       [       1999,      111000],\n",
              "       [       1995,     5500000],\n",
              "       [       1996,     4300000],\n",
              "       [       1997,    10000000],\n",
              "       [       1997,     7500000],\n",
              "       [       1997,     8800000],\n",
              "       [       1998,     7500000],\n",
              "       [       1999,     9500000],\n",
              "       [       1999,    13500000],\n",
              "       [       2000,    21000000],\n",
              "       [       2000,    21000000],\n",
              "       [       1999,    27400000],\n",
              "       [       1999,    21300000],\n",
              "       [       1999,    22000000],\n",
              "       [       2000,    42000000],\n",
              "       [       2001,   191000000],\n",
              "       [       2001,    45000000],\n",
              "       [       2002,    55000000],\n",
              "       [       2004,   112000000],\n",
              "       [       2004,   400000000],\n",
              "       [       2005,   169000000],\n",
              "       [       2006,   184000000],\n",
              "       [       2005,   228000000],\n",
              "       [       2006,   362000000],\n",
              "       [       2007,   540000000],\n",
              "       [       2008,    47000000],\n",
              "       [       2003,    54300000],\n",
              "       [       2003,   105900000],\n",
              "       [       2002,   220000000],\n",
              "       [       2005,   165000000],\n",
              "       [       2005,   250000000],\n",
              "       [       2006,   291000000],\n",
              "       [       2007,   169000000],\n",
              "       [       2003,   410000000],\n",
              "       [       2008,   600000000],\n",
              "       [       2009,   760000000],\n",
              "       [       2011,  1870000000],\n",
              "       [       2012,   432000000],\n",
              "       [       2007,   463000000],\n",
              "       [       2007,    26000000],\n",
              "       [       2008,   230000000],\n",
              "       [       2004,   592000000],\n",
              "       [       2007,   411000000],\n",
              "       [       2008,   731000000],\n",
              "       [       2008,   758000000],\n",
              "       [       2007,   789000000],\n",
              "       [       2009,   904000000],\n",
              "       [       2010,  1000000000],\n",
              "       [       2012,  2990000000],\n",
              "       [       2013,  1000000000],\n",
              "       [       2011,  1160000000],\n",
              "       [       2010,  1170000000],\n",
              "       [       2010,  1200000000],\n",
              "       [       2012,  1200000000],\n",
              "       [       2012,  1303000000],\n",
              "       [       2010,  1400000000],\n",
              "       [       2012,  1400000000],\n",
              "       [       2014,  1400000000],\n",
              "       [       2006,  1700000000],\n",
              "       [       2015,  1750000000],\n",
              "       [       2013,  1860000000],\n",
              "       [       2015,  1900000000],\n",
              "       [       2008,  1900000000],\n",
              "       [       2010,  2000000000],\n",
              "       [       2014,  2000000000],\n",
              "       [       2015,  2000000000],\n",
              "       [       2015,  3000000000],\n",
              "       [       2012,  2100000000],\n",
              "       [       2011,  2270000000],\n",
              "       [       2010,  2300000000],\n",
              "       [       2014,  2600000000],\n",
              "       [       2011,  2600000000],\n",
              "       [       2012,  2750000000],\n",
              "       [       2014,  3000000000],\n",
              "       [       2016,  3000000000],\n",
              "       [       2017,  5300000000],\n",
              "       [       2017,  5300000000],\n",
              "       [       2018,  8500000000],\n",
              "       [       2012,  3100000000],\n",
              "       [       2016,  3200000000],\n",
              "       [       2016,  3300000000],\n",
              "       [       2015,  3990000000],\n",
              "       [       2013,  4200000000],\n",
              "       [       2017,  4300000000],\n",
              "       [       2014,  4310000000],\n",
              "       [       2017,  4800000000],\n",
              "       [       2017,  4800000000],\n",
              "       [       2017,  4800000000],\n",
              "       [       2012,  5000000000],\n",
              "       [       2013,  5000000000],\n",
              "       [       2014,  5560000000],\n",
              "       [       2017,  6100000000],\n",
              "       [       2018,  6900000000],\n",
              "       [       2016,  4000000000],\n",
              "       [       2018,  6900000000],\n",
              "       [       2017,  5500000000],\n",
              "       [       2018,  5500000000],\n",
              "       [       2017,  7000000000],\n",
              "       [       2015,  7100000000],\n",
              "       [       2017,  8000000000],\n",
              "       [       2016,  7200000000],\n",
              "       [       2017,  8000000000],\n",
              "       [       2016,  8000000000],\n",
              "       [       2017,  9700000000],\n",
              "       [       2017,   250000000],\n",
              "       [       2015, 10000000000],\n",
              "       [       2017,  5450000000],\n",
              "       [       2018, 10000000000],\n",
              "       [       2017,  4300000000],\n",
              "       [       2017, 18000000000],\n",
              "       [       2017, 19200000000],\n",
              "       [       2018,  8876000000],\n",
              "       [       2018, 23600000000],\n",
              "       [       2018,  9000000000]])"
            ]
          },
          "metadata": {},
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Split up inputs (X) and targets (Y)\n",
        "\n",
        "X = data[:,0].reshape(-1, 1) # making an N x D matrix (hence X is 2D)\n",
        "Y = data[:,1]"
      ],
      "metadata": {
        "id": "MPUgWclMj9r8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot the data\n",
        "plt.scatter(X, Y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 463
        },
        "id": "wdOP_l3fkX7F",
        "outputId": "698af5ab-24fb-46f1-cf53-72e6b45d2a05"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.collections.PathCollection at 0x7e773049afb0>"
            ]
          },
          "metadata": {},
          "execution_count": 25
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Take the log since we want to have a linear model\n",
        "Y = np.log(Y)\n",
        "plt.scatter(X, Y);"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        },
        "id": "sodKNRX9kevC",
        "outputId": "81469215-2386-44a4-a895-873f595806c2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X = X - X.mean()"
      ],
      "metadata": {
        "id": "ORupkwnRk6jC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train the model\n",
        "model = tf.keras.models.Sequential([\n",
        "    tf.keras.layers.Input(shape=(1,)),      # Specifies the shape of input (D value)\n",
        "    tf.keras.layers.Dense(1)                # Output shape of 1 (there is 1 output)\n",
        "])"
      ],
      "metadata": {
        "id": "4awCSyx5lTb1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(\n",
        "    optimizer=tf.keras.optimizers.SGD(0.001, 0.9),    # used custom optimizers since data was not scaled\n",
        "    loss='mse'\n",
        ")"
      ],
      "metadata": {
        "id": "FMhFbdU9mJ3x"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def schedule(epoch, lr):\n",
        "  if epoch >= 50:\n",
        "    return 0.0001\n",
        "  return 0.001\n",
        "\n",
        "scheduler = tf.keras.callbacks.LearningRateScheduler(schedule)"
      ],
      "metadata": {
        "id": "usZDWTGqmmVL"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "r = model.fit(X, Y, epochs=200, callbacks=[scheduler])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Rf1OhPy0nEnp",
        "outputId": "8e9af0f1-e7dd-4c98-dc80-e42e54420042"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 343.7409 - lr: 0.0010\n",
            "Epoch 2/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 308.6382 - lr: 0.0010\n",
            "Epoch 3/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 252.9115 - lr: 0.0010\n",
            "Epoch 4/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 195.8343 - lr: 0.0010\n",
            "Epoch 5/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 166.1351 - lr: 0.0010\n",
            "Epoch 6/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 120.6484 - lr: 0.0010\n",
            "Epoch 7/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 86.0736 - lr: 0.0010\n",
            "Epoch 8/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 65.3504 - lr: 0.0010\n",
            "Epoch 9/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 74.6658 - lr: 0.0010\n",
            "Epoch 10/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 88.8734 - lr: 0.0010\n",
            "Epoch 11/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 62.0364 - lr: 0.0010\n",
            "Epoch 12/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 35.5026 - lr: 0.0010\n",
            "Epoch 13/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 24.7395 - lr: 0.0010\n",
            "Epoch 14/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 16.2153 - lr: 0.0010\n",
            "Epoch 15/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 8.4528 - lr: 0.0010\n",
            "Epoch 16/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 6.9156 - lr: 0.0010\n",
            "Epoch 17/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 5.0493 - lr: 0.0010\n",
            "Epoch 18/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 4.2639 - lr: 0.0010\n",
            "Epoch 19/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 3.7070 - lr: 0.0010\n",
            "Epoch 20/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 4.0231 - lr: 0.0010\n",
            "Epoch 21/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 3.0400 - lr: 0.0010\n",
            "Epoch 22/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 3.6021 - lr: 0.0010\n",
            "Epoch 23/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 2.3624 - lr: 0.0010\n",
            "Epoch 24/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 1.6270 - lr: 0.0010\n",
            "Epoch 25/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 2.1352 - lr: 0.0010\n",
            "Epoch 26/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 1.9349 - lr: 0.0010\n",
            "Epoch 27/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 1.5394 - lr: 0.0010\n",
            "Epoch 28/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 1.0043 - lr: 0.0010\n",
            "Epoch 29/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9568 - lr: 0.0010\n",
            "Epoch 30/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9335 - lr: 0.0010\n",
            "Epoch 31/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9437 - lr: 0.0010\n",
            "Epoch 32/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8912 - lr: 0.0010\n",
            "Epoch 33/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9217 - lr: 0.0010\n",
            "Epoch 34/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9024 - lr: 0.0010\n",
            "Epoch 35/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9003 - lr: 0.0010\n",
            "Epoch 36/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8785 - lr: 0.0010\n",
            "Epoch 37/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8850 - lr: 0.0010\n",
            "Epoch 38/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 1.1570 - lr: 0.0010\n",
            "Epoch 39/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 1.3170 - lr: 0.0010\n",
            "Epoch 40/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 1.1818 - lr: 0.0010\n",
            "Epoch 41/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 1.4296 - lr: 0.0010\n",
            "Epoch 42/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 1.2090 - lr: 0.0010\n",
            "Epoch 43/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 1.0755 - lr: 0.0010\n",
            "Epoch 44/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8830 - lr: 0.0010\n",
            "Epoch 45/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 1.6934 - lr: 0.0010\n",
            "Epoch 46/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 1.1343 - lr: 0.0010\n",
            "Epoch 47/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9298 - lr: 0.0010\n",
            "Epoch 48/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8925 - lr: 0.0010\n",
            "Epoch 49/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8858 - lr: 0.0010\n",
            "Epoch 50/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9030 - lr: 0.0010\n",
            "Epoch 51/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 1.1542 - lr: 1.0000e-04\n",
            "Epoch 52/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 1.2414 - lr: 1.0000e-04\n",
            "Epoch 53/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9066 - lr: 1.0000e-04\n",
            "Epoch 54/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9863 - lr: 1.0000e-04\n",
            "Epoch 55/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9562 - lr: 1.0000e-04\n",
            "Epoch 56/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8914 - lr: 1.0000e-04\n",
            "Epoch 57/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9056 - lr: 1.0000e-04\n",
            "Epoch 58/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8856 - lr: 1.0000e-04\n",
            "Epoch 59/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8734 - lr: 1.0000e-04\n",
            "Epoch 60/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8741 - lr: 1.0000e-04\n",
            "Epoch 61/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8746 - lr: 1.0000e-04\n",
            "Epoch 62/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8739 - lr: 1.0000e-04\n",
            "Epoch 63/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8742 - lr: 1.0000e-04\n",
            "Epoch 64/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8829 - lr: 1.0000e-04\n",
            "Epoch 65/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8921 - lr: 1.0000e-04\n",
            "Epoch 66/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8844 - lr: 1.0000e-04\n",
            "Epoch 67/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8798 - lr: 1.0000e-04\n",
            "Epoch 68/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8744 - lr: 1.0000e-04\n",
            "Epoch 69/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8733 - lr: 1.0000e-04\n",
            "Epoch 70/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8718 - lr: 1.0000e-04\n",
            "Epoch 71/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8783 - lr: 1.0000e-04\n",
            "Epoch 72/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8806 - lr: 1.0000e-04\n",
            "Epoch 73/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8729 - lr: 1.0000e-04\n",
            "Epoch 74/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8970 - lr: 1.0000e-04\n",
            "Epoch 75/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8943 - lr: 1.0000e-04\n",
            "Epoch 76/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8703 - lr: 1.0000e-04\n",
            "Epoch 77/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8717 - lr: 1.0000e-04\n",
            "Epoch 78/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8743 - lr: 1.0000e-04\n",
            "Epoch 79/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8776 - lr: 1.0000e-04\n",
            "Epoch 80/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8746 - lr: 1.0000e-04\n",
            "Epoch 81/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8744 - lr: 1.0000e-04\n",
            "Epoch 82/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8749 - lr: 1.0000e-04\n",
            "Epoch 83/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8688 - lr: 1.0000e-04\n",
            "Epoch 84/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8714 - lr: 1.0000e-04\n",
            "Epoch 85/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8720 - lr: 1.0000e-04\n",
            "Epoch 86/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8719 - lr: 1.0000e-04\n",
            "Epoch 87/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8734 - lr: 1.0000e-04\n",
            "Epoch 88/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8732 - lr: 1.0000e-04\n",
            "Epoch 89/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8765 - lr: 1.0000e-04\n",
            "Epoch 90/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8754 - lr: 1.0000e-04\n",
            "Epoch 91/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8743 - lr: 1.0000e-04\n",
            "Epoch 92/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8818 - lr: 1.0000e-04\n",
            "Epoch 93/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8727 - lr: 1.0000e-04\n",
            "Epoch 94/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8921 - lr: 1.0000e-04\n",
            "Epoch 95/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9346 - lr: 1.0000e-04\n",
            "Epoch 96/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8855 - lr: 1.0000e-04\n",
            "Epoch 97/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8790 - lr: 1.0000e-04\n",
            "Epoch 98/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8767 - lr: 1.0000e-04\n",
            "Epoch 99/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8770 - lr: 1.0000e-04\n",
            "Epoch 100/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8709 - lr: 1.0000e-04\n",
            "Epoch 101/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8708 - lr: 1.0000e-04\n",
            "Epoch 102/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8716 - lr: 1.0000e-04\n",
            "Epoch 103/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8734 - lr: 1.0000e-04\n",
            "Epoch 104/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8851 - lr: 1.0000e-04\n",
            "Epoch 105/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8804 - lr: 1.0000e-04\n",
            "Epoch 106/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8729 - lr: 1.0000e-04\n",
            "Epoch 107/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8757 - lr: 1.0000e-04\n",
            "Epoch 108/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8724 - lr: 1.0000e-04\n",
            "Epoch 109/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8722 - lr: 1.0000e-04\n",
            "Epoch 110/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9172 - lr: 1.0000e-04\n",
            "Epoch 111/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8879 - lr: 1.0000e-04\n",
            "Epoch 112/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8815 - lr: 1.0000e-04\n",
            "Epoch 113/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8888 - lr: 1.0000e-04\n",
            "Epoch 114/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8778 - lr: 1.0000e-04\n",
            "Epoch 115/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8724 - lr: 1.0000e-04\n",
            "Epoch 116/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8759 - lr: 1.0000e-04\n",
            "Epoch 117/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8774 - lr: 1.0000e-04\n",
            "Epoch 118/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8728 - lr: 1.0000e-04\n",
            "Epoch 119/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8707 - lr: 1.0000e-04\n",
            "Epoch 120/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8721 - lr: 1.0000e-04\n",
            "Epoch 121/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8751 - lr: 1.0000e-04\n",
            "Epoch 122/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8777 - lr: 1.0000e-04\n",
            "Epoch 123/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8724 - lr: 1.0000e-04\n",
            "Epoch 124/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8700 - lr: 1.0000e-04\n",
            "Epoch 125/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8724 - lr: 1.0000e-04\n",
            "Epoch 126/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8718 - lr: 1.0000e-04\n",
            "Epoch 127/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8737 - lr: 1.0000e-04\n",
            "Epoch 128/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8701 - lr: 1.0000e-04\n",
            "Epoch 129/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8720 - lr: 1.0000e-04\n",
            "Epoch 130/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8727 - lr: 1.0000e-04\n",
            "Epoch 131/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8732 - lr: 1.0000e-04\n",
            "Epoch 132/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8719 - lr: 1.0000e-04\n",
            "Epoch 133/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8729 - lr: 1.0000e-04\n",
            "Epoch 134/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8801 - lr: 1.0000e-04\n",
            "Epoch 135/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8760 - lr: 1.0000e-04\n",
            "Epoch 136/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8717 - lr: 1.0000e-04\n",
            "Epoch 137/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8681 - lr: 1.0000e-04\n",
            "Epoch 138/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8868 - lr: 1.0000e-04\n",
            "Epoch 139/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8776 - lr: 1.0000e-04\n",
            "Epoch 140/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8808 - lr: 1.0000e-04\n",
            "Epoch 141/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8775 - lr: 1.0000e-04\n",
            "Epoch 142/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8747 - lr: 1.0000e-04\n",
            "Epoch 143/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8772 - lr: 1.0000e-04\n",
            "Epoch 144/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8780 - lr: 1.0000e-04\n",
            "Epoch 145/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8769 - lr: 1.0000e-04\n",
            "Epoch 146/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8736 - lr: 1.0000e-04\n",
            "Epoch 147/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8711 - lr: 1.0000e-04\n",
            "Epoch 148/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8715 - lr: 1.0000e-04\n",
            "Epoch 149/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8776 - lr: 1.0000e-04\n",
            "Epoch 150/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.8941 - lr: 1.0000e-04\n",
            "Epoch 151/200\n",
            "6/6 [==============================] - 0s 2ms/step - loss: 0.9232 - lr: 1.0000e-04\n",
            "Epoch 152/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8844 - lr: 1.0000e-04\n",
            "Epoch 153/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8745 - lr: 1.0000e-04\n",
            "Epoch 154/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8745 - lr: 1.0000e-04\n",
            "Epoch 155/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8721 - lr: 1.0000e-04\n",
            "Epoch 156/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8781 - lr: 1.0000e-04\n",
            "Epoch 157/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8761 - lr: 1.0000e-04\n",
            "Epoch 158/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8748 - lr: 1.0000e-04\n",
            "Epoch 159/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8908 - lr: 1.0000e-04\n",
            "Epoch 160/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8779 - lr: 1.0000e-04\n",
            "Epoch 161/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8942 - lr: 1.0000e-04\n",
            "Epoch 162/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.9111 - lr: 1.0000e-04\n",
            "Epoch 163/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8820 - lr: 1.0000e-04\n",
            "Epoch 164/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8750 - lr: 1.0000e-04\n",
            "Epoch 165/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8758 - lr: 1.0000e-04\n",
            "Epoch 166/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8708 - lr: 1.0000e-04\n",
            "Epoch 167/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8774 - lr: 1.0000e-04\n",
            "Epoch 168/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8829 - lr: 1.0000e-04\n",
            "Epoch 169/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8737 - lr: 1.0000e-04\n",
            "Epoch 170/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8748 - lr: 1.0000e-04\n",
            "Epoch 171/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8817 - lr: 1.0000e-04\n",
            "Epoch 172/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8775 - lr: 1.0000e-04\n",
            "Epoch 173/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8693 - lr: 1.0000e-04\n",
            "Epoch 174/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8723 - lr: 1.0000e-04\n",
            "Epoch 175/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8766 - lr: 1.0000e-04\n",
            "Epoch 176/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8769 - lr: 1.0000e-04\n",
            "Epoch 177/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8710 - lr: 1.0000e-04\n",
            "Epoch 178/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8793 - lr: 1.0000e-04\n",
            "Epoch 179/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8830 - lr: 1.0000e-04\n",
            "Epoch 180/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8734 - lr: 1.0000e-04\n",
            "Epoch 181/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8784 - lr: 1.0000e-04\n",
            "Epoch 182/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8872 - lr: 1.0000e-04\n",
            "Epoch 183/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8727 - lr: 1.0000e-04\n",
            "Epoch 184/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8863 - lr: 1.0000e-04\n",
            "Epoch 185/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8802 - lr: 1.0000e-04\n",
            "Epoch 186/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8740 - lr: 1.0000e-04\n",
            "Epoch 187/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8866 - lr: 1.0000e-04\n",
            "Epoch 188/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8801 - lr: 1.0000e-04\n",
            "Epoch 189/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8804 - lr: 1.0000e-04\n",
            "Epoch 190/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8732 - lr: 1.0000e-04\n",
            "Epoch 191/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8715 - lr: 1.0000e-04\n",
            "Epoch 192/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8701 - lr: 1.0000e-04\n",
            "Epoch 193/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8752 - lr: 1.0000e-04\n",
            "Epoch 194/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8728 - lr: 1.0000e-04\n",
            "Epoch 195/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8778 - lr: 1.0000e-04\n",
            "Epoch 196/200\n",
            "6/6 [==============================] - 0s 4ms/step - loss: 0.8758 - lr: 1.0000e-04\n",
            "Epoch 197/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8795 - lr: 1.0000e-04\n",
            "Epoch 198/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8705 - lr: 1.0000e-04\n",
            "Epoch 199/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8780 - lr: 1.0000e-04\n",
            "Epoch 200/200\n",
            "6/6 [==============================] - 0s 3ms/step - loss: 0.8705 - lr: 1.0000e-04\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot the loss\n",
        "plt.plot(r.history['loss'], label='loss')\n",
        "plt.legend();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        },
        "id": "n7SOJpYFnYYJ",
        "outputId": "a62e8948-ae77-43dd-dc70-1096237b7eaf"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.layers"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5a-nHKh7niTq",
        "outputId": "65cb2a5a-2993-440f-8056-6b546577cfb7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<keras.layers.core.dense.Dense at 0x7e77305580d0>]"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.layers[0].get_weights()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZfNoZeGdn36c",
        "outputId": "b7b70e1e-afbd-4abe-bdaf-d7bf8a94f8eb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[array([[0.34102172]], dtype=float32), array([17.76452], dtype=float32)]"
            ]
          },
          "metadata": {},
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "a = model.layers[0].get_weights()[0][0,0]"
      ],
      "metadata": {
        "id": "BIQjXj5Dn-q5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "a"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pslWCDYGoVXC",
        "outputId": "2efd8430-fa27-4311-d30d-3f43ee4678a7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.34102172"
            ]
          },
          "metadata": {},
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"Time to double\", np.log(2)/a)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xVlstWILoWHl",
        "outputId": "3735bb7f-24cb-4c3c-b693-4fccba258418"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Time to double 2.0325602354050374\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Optional\n",
        "\n",
        "X = X.flatten()\n",
        "denominator = X.dot(X) - X.mean() * X.sum()\n",
        "a = ( X.dot(Y) - Y.mean() * X.sum() ) / denominator\n",
        "b = ( Y.mean() * X.dot(X) - X.mean() * X.dot(Y) ) / denominator\n",
        "print(a, b)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AADeoaQBpm_-",
        "outputId": "a88eebb3-7cc0-43c8-f1bc-d8944e447189"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.341682487387313 17.764939393631764\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Part 2: Making Predictions**\n",
        "This goes with the lecture \"Making Predictions\""
      ],
      "metadata": {
        "id": "pZFdTGu9cdah"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Make sure the line fits our data\n",
        "Y_hat = model.predict(X).flatten()\n",
        "plt.scatter(X, Y)\n",
        "plt.plot(X, Y_hat)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 466
        },
        "id": "fus5xaDycehc",
        "outputId": "09c2fafe-3f34-459b-d879-da90f100dd19"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "6/6 [==============================] - 0s 3ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7e772296ac50>]"
            ]
          },
          "metadata": {},
          "execution_count": 39
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Manual calculation\n",
        "\n",
        "# Get the weights\n",
        "w, b = model.layers[0].get_weights()\n",
        "\n",
        "# Reshape X because we flattened it again earlier \\\n",
        "X = X.reshape(-1, 1)\n",
        "\n",
        "# (N x 1) x (1 x 1) + (1) --> (N x 1)\n",
        "Y_hat2 = (X.dot(w) + b).flatten()\n",
        "\n",
        "# Don't use == for floating points\n",
        "np.allclose(Y_hat, Y_hat2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KwsuSxLocxcA",
        "outputId": "ba228167-6e18-4d47-c13b-5211e0867a18"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {},
          "execution_count": 40
        }
      ]
    }
  ]
}