Simple_Regression_Model.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Simple Regression Model",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ondaPeBrRPQJ",
        "outputId": "0b83272a-c39a-46d7-e691-a9eafba55925"
      },
      "source": [
        "%tensorflow_version 1.x"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "TensorFlow 1.x selected.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e8QYXmEURfDw"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline \n",
        "import tensorflow as tf"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i1jcxd9WZe4h"
      },
      "source": [
        "## Creating the Dataset\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8Mv6yEPZYcvV"
      },
      "source": [
        "x_data = np.linspace(0.0,10.0,1000000)"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Y9HMQZCqFOZ6"
      },
      "source": [
        "noise = np.random.randn(len(x_data))"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nUrnzXSzFcoB"
      },
      "source": [
        " #y = mx + b\n",
        "b = 5\n",
        "\n",
        "y_true =  (0.5 * x_data ) + 5 + noise"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8x9BmxLlFjcW"
      },
      "source": [
        "x_df = pd.DataFrame(data=x_data, columns=['X Data'])\n",
        "y_df = pd.DataFrame(data=y_true, columns=['Y Data'])"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "T5UOhU2lGHeD",
        "outputId": "b0ade9f2-c1ee-4f56-afbf-21e6e8e12061"
      },
      "source": [
        "x_df.head()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>X Data</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.00001</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.00002</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.00003</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.00004</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    X Data\n",
              "0  0.00000\n",
              "1  0.00001\n",
              "2  0.00002\n",
              "3  0.00003\n",
              "4  0.00004"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "fg6EF9v9GsPb",
        "outputId": "b8d81d9a-65e7-4972-dea6-62dd8efc08df"
      },
      "source": [
        "y_df.head()"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Y Data</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>4.100354</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>6.279671</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>6.071362</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3.768398</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5.324067</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     Y Data\n",
              "0  4.100354\n",
              "1  6.279671\n",
              "2  6.071362\n",
              "3  3.768398\n",
              "4  5.324067"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VV9HtwvgGuVL"
      },
      "source": [
        "my_data = pd.concat([x_df,y_df],axis=1)"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "JqTJMp4MGyeh",
        "outputId": "a6790b0c-05f0-4005-f690-76be1462617a"
      },
      "source": [
        "my_data.head()"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>X Data</th>\n",
              "      <th>Y Data</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.00000</td>\n",
              "      <td>4.100354</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.00001</td>\n",
              "      <td>6.279671</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.00002</td>\n",
              "      <td>6.071362</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.00003</td>\n",
              "      <td>3.768398</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.00004</td>\n",
              "      <td>5.324067</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    X Data    Y Data\n",
              "0  0.00000  4.100354\n",
              "1  0.00001  6.279671\n",
              "2  0.00002  6.071362\n",
              "3  0.00003  3.768398\n",
              "4  0.00004  5.324067"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "NM8nNqyXHZjP",
        "outputId": "2ce38217-04bc-4233-9d9a-201bb2ffe090"
      },
      "source": [
        "my_data.sample(n=250).plot(kind='scatter', x='X Data', y='Y Data')"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f039af0a910>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Dl5ZVLKeZkNJ"
      },
      "source": [
        "##Creating the Optimizer"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3J3ip1qIQNyn"
      },
      "source": [
        "batch_size = 8"
      ],
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vqyS1F7NHxMB",
        "outputId": "4ab855e6-1170-4dbd-df5e-58d5204ee4b0"
      },
      "source": [
        "np.random.randn(2)"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 1.00497378, -1.17179002])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GjSiL9esQF5y"
      },
      "source": [
        "m = tf.Variable(0.5)\n",
        "b = tf.Variable(1.0)"
      ],
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rSa-Y_mFQQ2a"
      },
      "source": [
        "xph = tf.placeholder(tf.float32,[batch_size])\n",
        "yph = tf.placeholder(tf.float32,[batch_size])"
      ],
      "execution_count": 41,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1JiucGiWQaIO"
      },
      "source": [
        "y_pred = m*xph+b"
      ],
      "execution_count": 42,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oO-lmTliQheB"
      },
      "source": [
        "error = tf.reduce_sum(tf.square(yph-y_pred))"
      ],
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9uWOgXnpQu8q"
      },
      "source": [
        "optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.001)\n",
        "train = optimizer.minimize(error)"
      ],
      "execution_count": 44,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fVxSwCJkQ-mn"
      },
      "source": [
        "init = tf.global_variables_initializer()"
      ],
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RzHbyX7aZpHJ"
      },
      "source": [
        "##Training the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k9yvO8CcUUup"
      },
      "source": [
        "with tf.Session() as sess:\n",
        "    \n",
        "    sess.run(init)\n",
        "    \n",
        "    batches = 1000\n",
        "    \n",
        "    for i in range(batches):\n",
        "        \n",
        "        rand_ind = np.random.randint(len(x_data),size=batch_size)\n",
        "        \n",
        "        feed = {xph:x_data[rand_ind],yph:y_true[rand_ind]}\n",
        "        \n",
        "        sess.run(train,feed_dict=feed)\n",
        "        \n",
        "    model_m,model_b = sess.run([m,b])   "
      ],
      "execution_count": 50,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gb4qU4xfZro5"
      },
      "source": [
        "##Printing the data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dXcZn9TaZ5da",
        "outputId": "070f51d7-5b3c-43e2-8748-63818ad7132c"
      },
      "source": [
        "model_m"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.53795433"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RJ-l8o6yZ7x5",
        "outputId": "608251df-ca36-4cf7-ba5d-899872361f82"
      },
      "source": [
        "model_b"
      ],
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "4.8801346"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "id": "OlnSDeSmRVlV",
        "outputId": "f7d6e101-08ae-4842-dfcc-4fee02ab24db"
      },
      "source": [
        "y_final = m_model*x_data + b_model\n",
        "my_data.sample(n=250).plot(kind='scatter', x='X Data', y='Y Data')\n",
        "plt.plot(x_data,y_final,'r')"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f0393fc4a10>]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 53
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IQXkLlTcT5fN"
      },
      "source": [
        ""
      ],
      "execution_count": 23,
      "outputs": []
    }
  ]
}