Module 6 Regression Lab 2.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "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/Nitin-Sagar-B/FMML_PROJECTS_AND_LABS/blob/main/Module%206%20Regression%20Lab%202.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "H-Cp5UaReiof"
      },
      "source": [
        "#Regression Lab 2: Loss Functions\n"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "```\n",
        "Module Coordinator : Nikunj Nawal\n",
        "```\n"
      ],
      "metadata": {
        "id": "TSBLUen_aWef"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rbxg8WQcetL2"
      },
      "source": [
        "## Motivation\n",
        "\n",
        "The main goal of machine learning tasks is to \"learn\" something about data we feed to our models. But how does the model actually learn anything? To understand this, we need to first understand what objective functions are.\n",
        "\n",
        "Mathematically, objective functions are a function whose values we desire to minimize/maximize. In the context of machine learning, the objective is commonly referred to as a **loss function**. Loss functions indicate how well the model is doing on the dataset.\n",
        "\n",
        "The value of the objective function is simply called **loss**. Our goal is to eventually find the best model parameters that has the least average loss on the data after training.\n",
        "\n",
        "\n",
        "We are familiar with the idea of **error** from the linear regression lab. RMSE is a common loss function used in machine learning. However, there are many other loss functions employed, depending on what the task is (classification, regression etc.). In this lab, we will take a look at some common loss functions, and understand why they are used where they are.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Q_TjX0v1Upyt"
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "plt.style.use('ggplot')\n",
        "\n",
        "# interactive visualization"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "o9lGgMhd9hAX"
      },
      "source": [
        "# 1. Regression\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f9qiER5ef8Sh",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        },
        "outputId": "a324fb94-b645-47ad-b035-b12487c6f47d"
      },
      "source": [
        "# initializing random data\n",
        "\n",
        "np.random.seed(0)\n",
        "x = np.linspace(0,100,100)\n",
        "y_actual = np.copy(x)\n",
        "y_pred = x + np.random.randint(-10,10,100)\n",
        "\n",
        "plt.plot(x,y_actual,color=\"red\")\n",
        "plt.scatter(x,y_pred,color=\"blue\")\n",
        "plt.show()"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JAOKu9cGhf7t"
      },
      "source": [
        "### Mean Squared Error\n",
        "\n",
        "As we have seen before, the formula for MSE is\n",
        "\n",
        "$MSE = \\frac{1}{m} \\sum_{i=1}^n (y-y_p)^2$\n",
        "\n",
        "The image below depicts a visualization of what the squared error is.\n",
        "\n",
        "<img src=\"https://drive.google.com/uc?export=view&id=12jmqRrrqknIWKhwRpgRIJAErBjkEcyx9\" width=\"350\" height=\"350\" align=\"middle\"/>\n",
        "\n",
        "MSE is also referred to as L2 loss.\n",
        "\n",
        "Typically, we would want the units of error to be the same as the data, so we use Root Mean Squared Error instead.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "y0KIs1xP2AbG",
        "outputId": "6c5ed262-01a2-4475-8f65-776a7b7e1890"
      },
      "source": [
        "mse = np.mean((y_actual-y_pred)**2)\n",
        "print(\"MSE is: {}\".format(mse))\n",
        "print(\"RMSE is: {}\".format(np.sqrt(mse)))"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MSE is: 38.68\n",
            "RMSE is: 6.219324722186485\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "05Gx5pbfhk3P"
      },
      "source": [
        "### Mean Absolute Error\n",
        "\n",
        "$MAE = \\frac{1}{m} \\sum_{i=1}^n |y-y_p|$\n",
        "\n",
        "MAE is also commonly referred to as L1 loss."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PfqvIN2ThjWO",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f7918b13-8259-4ec8-f749-cafba25c4bdd"
      },
      "source": [
        "mae = np.mean(np.abs(y_actual-y_pred))\n",
        "print(\"MAE is: {}\".format(mae))"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MAE is: 5.5\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4j7VlwmH4WVU"
      },
      "source": [
        "MSE penalizes the model for making large errors by squaring the difference. However, this also means that MSE cannot handle outliers well, since they would throw the model off.\n",
        "\n",
        "On the other hand, MAE is robust to outliers. But, MAE is non-differentiable, making it difficult to perform mathematical operations on it.\n",
        "\n",
        "Hence, RMSE is most commonly used as a metric for evaluation."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "buXwWqWjhT_9"
      },
      "source": [
        "# 2. Classification\n",
        "\n",
        "Classification tasks involve some data, along with labels assigned to that data. For example, we may have pictures of cats and dogs, and the corresponding labels assigned to those images. We then want to train a model that can predict a label for some input image belonging to either label class (cat or dog)."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qiJUcm4vhthC"
      },
      "source": [
        "## The concept of maximum likelihood\n",
        "\n",
        "Ideally, we would want to obtain a model $\\hat{\\theta}$ which maximizes the probability of correctly predicting data. Mathematically, the likelihood is:\n",
        "\n",
        "$ \\prod_{i=1}^{N}  \\: \\hat{\\theta_i}^{\\theta_i}$\n",
        "\n",
        "This is because our actual data ($\\theta_i$) is either 0 or 1 depending on the label of the data, and our predicted data ($\\hat{\\theta_i}$) is a probability.\n",
        "\n",
        "We usually consider the negative log of the likelihood, since log is monotonic and easier to deal with as an optimization problem.\n",
        "\n",
        "Hence, we have\n",
        "\n",
        "$MLE = $ arg min $\\: - \\sum_{i=1}^{N} \\theta_i \\: log \\: \\hat{\\theta_i}$\n",
        "\n",
        "\\\n",
        "\n",
        "\n",
        "## Entropy and Cross-Entropy\n",
        "\n",
        "For some event $x$, we can obtain **information** based on its probability distribution. For example, for some event that occurs with $p=1$, we gain no information. Now, if we flip a coin and see that we obtained heads, we say that we got 1 bit of information.\n",
        "\n",
        "Thus, the information $I$ of some event $x$ is:\n",
        "\n",
        "$I = -log_2 \\: p(x)$\n",
        "\n",
        "We say that events with a low probability of occuring give high information, and those with a low probability give low information.\n",
        "\n",
        "Now, if we have some random variable $X$, its **entropy** is the expected value of the information obtained.\n",
        "\n",
        "$H(x) = - \\sum_{k} p_k \\: log \\: p_k$\n",
        "\n",
        "Now, suppose we are trying to correctly predict labels of some data. Let P be the true distribution of the labels, and Q be the predicted distribution of labels.\n",
        "\n",
        "Cross-Entropy is then defined as:\n",
        "\n",
        "$H(P,Q) = - \\sum_{x} P(x)\\: log \\: Q(x)$\n",
        "\n",
        "which is basically the entropy between two probability distributions over the same events.\n",
        "\n",
        "\\\n",
        "\n",
        "\n",
        "## MLE and Cross-Entropy\n",
        "\n",
        "So, in the case of classification, the equation for minimization of the cross-entropy between the actual data and the predicted data would be:\n",
        "\n",
        "arg min $\\: -\\sum_{i=1}^{N} P(x) \\: log \\: Q(x)$\n",
        "\n",
        "$= $ arg min $\\: -\\sum_{i=1}^{N} \\theta_i \\: log \\: \\hat{\\theta_i}$\n",
        "\n",
        "\n",
        "which is exactly the same as what we had obtained for minimizing the negative log likelihood. Hence, the two problems are equivalent."
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "1/1+e^(-x)"
      ],
      "metadata": {
        "id": "UyJN_UiEbbch"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mLKuJyHth5-4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 607
        },
        "outputId": "33cc04c4-5076-43ec-8ea1-3924f85179fe"
      },
      "source": [
        "def sigmoid(z):\n",
        "  return 1/(1 + np.exp(-z))\n",
        "\n",
        "def cross_entropy_loss(y_,y):\n",
        "  if y==1:\n",
        "    return -np.log(y_)\n",
        "\n",
        "  return -np.log(1-y_)\n",
        "\n",
        "z = np.arange(-10, 10, 0.1)\n",
        "h_z = sigmoid(z)\n",
        "\n",
        "cost_1 = cross_entropy_loss(h_z, 1)\n",
        "cost_0 = cross_entropy_loss(h_z, 0)\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(8,6))\n",
        "plt.plot(h_z, cost_1, label='y=1')\n",
        "plt.plot(h_z, cost_0, label='y=0')\n",
        "plt.xlabel('predicted probability')\n",
        "plt.ylabel('loss')\n",
        "plt.legend(loc='best')\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x600 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## KNN-regression\n",
        "\n",
        "Till now, we have studied KNN as a classification algorithm, which uses feature similarity and voting to decide the class labels of unseen data.\n",
        "\n",
        "For regression, we use the same idea: We take the y-values of the k nearest neighbours, and compute their average, which is our predicted y-value. We do this for every x whose value we want to predict.\n",
        "\n",
        "In the visualization below, we have k=3, and the red dots are the new x-values whose output we want to predict.\n",
        "\n",
        "![image.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "f42L9ZmuWVIQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn import neighbors\n",
        "from sklearn.metrics import mean_squared_error\n",
        "from math import sqrt\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "import numpy as np"
      ],
      "metadata": {
        "id": "cQIX2HoAbqqD"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_path = \"https://raw.githubusercontent.com/PranavTadimeti/Regression-lab2/main/BigMart-train.csv\"\n",
        "\n",
        "train = pd.read_csv(train_path)"
      ],
      "metadata": {
        "id": "VLMVfcylWlWB"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Preprocess Data\n",
        "\n",
        "def preprocess(df):\n",
        "  mean_weight = df['Item_Weight'].mean()\n",
        "  df['Item_Weight'].fillna(mean_weight,inplace=True)\n",
        "\n",
        "  mode_size = df['Outlet_Size'].mode()\n",
        "  df['Outlet_Size'].fillna(mode_size,inplace=True)\n",
        "\n",
        "  df.drop(['Item_Identifier','Outlet_Identifier'],axis=1,inplace=True)\n",
        "\n",
        "  return pd.get_dummies(df)\n",
        "\n",
        "train = preprocess(train)\n",
        "\n",
        "train.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 273
        },
        "id": "bkMzDwYaYTlQ",
        "outputId": "d42e5960-9036-4a6c-9d3f-a33217432c68"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   Item_Weight  Item_Visibility  Item_MRP  Outlet_Establishment_Year  \\\n",
              "0         9.30         0.016047  249.8092                       1999   \n",
              "1         5.92         0.019278   48.2692                       2009   \n",
              "2        17.50         0.016760  141.6180                       1999   \n",
              "3        19.20         0.000000  182.0950                       1998   \n",
              "4         8.93         0.000000   53.8614                       1987   \n",
              "\n",
              "   Item_Outlet_Sales  Item_Fat_Content_LF  Item_Fat_Content_Low Fat  \\\n",
              "0          3735.1380                    0                         1   \n",
              "1           443.4228                    0                         0   \n",
              "2          2097.2700                    0                         1   \n",
              "3           732.3800                    0                         0   \n",
              "4           994.7052                    0                         1   \n",
              "\n",
              "   Item_Fat_Content_Regular  Item_Fat_Content_low fat  Item_Fat_Content_reg  \\\n",
              "0                         0                         0                     0   \n",
              "1                         1                         0                     0   \n",
              "2                         0                         0                     0   \n",
              "3                         1                         0                     0   \n",
              "4                         0                         0                     0   \n",
              "\n",
              "   ...  Outlet_Size_High  Outlet_Size_Medium  Outlet_Size_Small  \\\n",
              "0  ...                 0                   1                  0   \n",
              "1  ...                 0                   1                  0   \n",
              "2  ...                 0                   1                  0   \n",
              "3  ...                 0                   0                  0   \n",
              "4  ...                 1                   0                  0   \n",
              "\n",
              "   Outlet_Location_Type_Tier 1  Outlet_Location_Type_Tier 2  \\\n",
              "0                            1                            0   \n",
              "1                            0                            0   \n",
              "2                            1                            0   \n",
              "3                            0                            0   \n",
              "4                            0                            0   \n",
              "\n",
              "   Outlet_Location_Type_Tier 3  Outlet_Type_Grocery Store  \\\n",
              "0                            0                          0   \n",
              "1                            1                          0   \n",
              "2                            0                          0   \n",
              "3                            1                          1   \n",
              "4                            1                          0   \n",
              "\n",
              "   Outlet_Type_Supermarket Type1  Outlet_Type_Supermarket Type2  \\\n",
              "0                              1                              0   \n",
              "1                              0                              1   \n",
              "2                              1                              0   \n",
              "3                              0                              0   \n",
              "4                              1                              0   \n",
              "\n",
              "   Outlet_Type_Supermarket Type3  \n",
              "0                              0  \n",
              "1                              0  \n",
              "2                              0  \n",
              "3                              0  \n",
              "4                              0  \n",
              "\n",
              "[5 rows x 36 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-0ab425f6-8bbb-48ab-b17a-7e7eb5dca39f\" class=\"colab-df-container\">\n",
              "    <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>Item_Weight</th>\n",
              "      <th>Item_Visibility</th>\n",
              "      <th>Item_MRP</th>\n",
              "      <th>Outlet_Establishment_Year</th>\n",
              "      <th>Item_Outlet_Sales</th>\n",
              "      <th>Item_Fat_Content_LF</th>\n",
              "      <th>Item_Fat_Content_Low Fat</th>\n",
              "      <th>Item_Fat_Content_Regular</th>\n",
              "      <th>Item_Fat_Content_low fat</th>\n",
              "      <th>Item_Fat_Content_reg</th>\n",
              "      <th>...</th>\n",
              "      <th>Outlet_Size_High</th>\n",
              "      <th>Outlet_Size_Medium</th>\n",
              "      <th>Outlet_Size_Small</th>\n",
              "      <th>Outlet_Location_Type_Tier 1</th>\n",
              "      <th>Outlet_Location_Type_Tier 2</th>\n",
              "      <th>Outlet_Location_Type_Tier 3</th>\n",
              "      <th>Outlet_Type_Grocery Store</th>\n",
              "      <th>Outlet_Type_Supermarket Type1</th>\n",
              "      <th>Outlet_Type_Supermarket Type2</th>\n",
              "      <th>Outlet_Type_Supermarket Type3</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>9.30</td>\n",
              "      <td>0.016047</td>\n",
              "      <td>249.8092</td>\n",
              "      <td>1999</td>\n",
              "      <td>3735.1380</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5.92</td>\n",
              "      <td>0.019278</td>\n",
              "      <td>48.2692</td>\n",
              "      <td>2009</td>\n",
              "      <td>443.4228</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>17.50</td>\n",
              "      <td>0.016760</td>\n",
              "      <td>141.6180</td>\n",
              "      <td>1999</td>\n",
              "      <td>2097.2700</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>19.20</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>182.0950</td>\n",
              "      <td>1998</td>\n",
              "      <td>732.3800</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>8.93</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>53.8614</td>\n",
              "      <td>1987</td>\n",
              "      <td>994.7052</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 36 columns</p>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-0ab425f6-8bbb-48ab-b17a-7e7eb5dca39f')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-0ab425f6-8bbb-48ab-b17a-7e7eb5dca39f button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-0ab425f6-8bbb-48ab-b17a-7e7eb5dca39f');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-abfd7a95-91c4-4d9e-b55c-249a46c002ba\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-abfd7a95-91c4-4d9e-b55c-249a46c002ba')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-abfd7a95-91c4-4d9e-b55c-249a46c002ba button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train , test = train_test_split(train, test_size = 0.3)\n",
        "\n",
        "x_train = train.drop('Item_Outlet_Sales', axis=1)\n",
        "y_train = train['Item_Outlet_Sales']\n",
        "\n",
        "x_test = test.drop('Item_Outlet_Sales', axis = 1)\n",
        "y_test = test['Item_Outlet_Sales']"
      ],
      "metadata": {
        "id": "JyYJSSj-Z3mT"
      },
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "scaler = MinMaxScaler(feature_range=(0, 1))\n",
        "\n",
        "x_train_scaled = scaler.fit_transform(x_train)\n",
        "x_train = pd.DataFrame(x_train_scaled)\n",
        "\n",
        "x_test_scaled = scaler.fit_transform(x_test)\n",
        "x_test = pd.DataFrame(x_test_scaled)"
      ],
      "metadata": {
        "id": "Pp38lIDgZsyn"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def knn_regression(k,x_train,y_train,x_test,y_test):\n",
        "    model = neighbors.KNeighborsRegressor(n_neighbors = k)\n",
        "\n",
        "    model.fit(x_train, y_train)\n",
        "    pred=model.predict(x_test)\n",
        "    error = sqrt(mean_squared_error(y_test,pred))\n",
        "    rmse_val.append(error)"
      ],
      "metadata": {
        "id": "XM_GDnVqbVKI"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "rmse_val = []\n",
        "\n",
        "for k in range(1,21):\n",
        "  knn_regression(k,x_train,y_train,x_test,y_test)\n",
        "\n",
        "figure = plt.figure(figsize=(10,10))\n",
        "plt.plot(np.arange(1,21),rmse_val)\n",
        "plt.xlabel(\"K\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.show()\n",
        "\n",
        "rmse_val = np.asarray(rmse_val)\n",
        "print(\"Minimum error {} is at k = {}\".format(np.min(rmse_val),np.argmin(rmse_val)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 871
        },
        "id": "P0rIZOClcKI6",
        "outputId": "0d892de0-f938-4983-dad4-6a47b13ff946"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x1000 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Minimum error 1278.5816281094171 is at k = 6\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Exercises\n",
        "\n",
        "1. Try out these loss functions on regression tasks, and try to understand the model performance based on the loss function.\n",
        "2. Explore other loss functions and try to understand when and why they are used.\n",
        "3. Try out KNN-regression on other datasets see which values of K give the best results."
      ],
      "metadata": {
        "id": "0QNjfdMXk2AH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "<h1>Excercise - 1"
      ],
      "metadata": {
        "id": "IW53oRu5wIFT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Excercise 1\n",
        "from sklearn.metrics import mean_absolute_error, mean_squared_error\n",
        "from sklearn.linear_model import LinearRegression\n",
        "\n",
        "# Train a linear regression model\n",
        "model = LinearRegression()\n",
        "model.fit(x_train, y_train)\n",
        "\n",
        "# Make predictions\n",
        "y_pred = model.predict(x_test)\n",
        "\n",
        "# Calculate loss functions\n",
        "mse = mean_squared_error(y_test, y_pred)\n",
        "rmse = np.sqrt(mse)\n",
        "mae = mean_absolute_error(y_test, y_pred)\n",
        "\n",
        "print(f\"MSE: {mse}\")\n",
        "print(f\"RMSE: {rmse}\")\n",
        "print(f\"MAE: {mae}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IWl1G4oqtpfm",
        "outputId": "ed2c124e-ea13-4c02-ab1d-0cc8e6d6f4f0"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "MSE: 1381770.8086451201\n",
            "RMSE: 1175.487477026072\n",
            "MAE: 872.8599159953069\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "<h1>Excercise - 2"
      ],
      "metadata": {
        "id": "8Ii7Qnq7wRv6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.linear_model import HuberRegressor\n",
        "\n",
        "# Train a Huber regressor\n",
        "model = HuberRegressor()\n",
        "model.fit(x_train, y_train)\n",
        "\n",
        "# Make predictions\n",
        "y_pred = model.predict(x_test)\n",
        "\n",
        "# Calculate Huber loss\n",
        "huber_loss = model.score(x_test, y_test)\n",
        "print(f\"Huber Loss: {huber_loss}\")\n",
        "\n",
        "def mean_absolute_percentage_error(y_true, y_pred):\n",
        "    return np.mean(np.abs((y_true - y_pred) / y_true)) * 100\n",
        "\n",
        "# Calculate MAPE\n",
        "mape = mean_absolute_percentage_error(y_test, y_pred)\n",
        "print(f\"MAPE on given dataset: {mape}%\")\n",
        "\n",
        "from sklearn.metrics import r2_score\n",
        "\n",
        "# Calculate R^2 Score\n",
        "r2 = r2_score(y_test, y_pred)\n",
        "print(f\"R^2 Score on given dataset: {r2}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Zg9FJnZquTxb",
        "outputId": "bb4075ff-91c6-427d-ba2f-5403a0b4fef0"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Huber Loss: 0.5497969496598321\n",
            "MAPE on given dataset: 96.41404661259409%\n",
            "R^2 Score on given dataset: 0.5497969496598321\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "What I understood and observed about these loss functions:\n",
        "\n",
        "1. **Huber Loss**: I observed that the Huber Loss is a loss function utilized in robust regression. It's less sensitive to outliers in data compared to the squared error loss. This function combines the properties of both quadratic loss (squared error) and absolute loss (absolute error), providing a more robust solution. It's typically used when there's a need to avoid assigning excessive weight to outliers.\n",
        "\n",
        "2. **Mean Absolute Percentage Error (MAPE)**: I understand that the Mean Absolute Percentage Error (MAPE) is a statistical measure used to determine the accuracy of a forecasting method. It calculates the average of the absolute percentage errors of each entry in a dataset. This helps in determining how accurate the forecasted quantities were in comparison with the actual quantities. It's often used due to its ease of interpretation. For instance, a MAPE value of 14% implies that the average difference between the forecasted value and the actual value is 14%.\n",
        "\n",
        "3. **R^2 Score**: Also known as the coefficient of determination, the R^2 score measures the predictive power of a statistical model. The value of R^2 can range from 0 to 1. An R^2 of 1 indicates that the regression predictions perfectly fit the data. Conversely, an R^2 of 0 suggests that the data is random or cannot be fit to the model."
      ],
      "metadata": {
        "id": "EFTLhyK8xgGe"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "<h1>Excercise - 3"
      ],
      "metadata": {
        "id": "fkeQ3fg4ymit"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn import datasets\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from sklearn import neighbors\n",
        "from math import sqrt\n",
        "from sklearn.metrics import mean_squared_error\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "\n",
        "# Load the Iris dataset\n",
        "iris = datasets.load_iris()\n",
        "iris_data = iris.data\n",
        "iris_target = iris.target\n",
        "\n",
        "# Split the data into train and test sets\n",
        "x_train, x_test, y_train, y_test = train_test_split(iris_data, iris_target, test_size = 0.3, random_state = 1)\n",
        "\n",
        "# Scale the features\n",
        "scaler = MinMaxScaler(feature_range=(0, 1))\n",
        "\n",
        "x_train_scaled = scaler.fit_transform(x_train)\n",
        "x_train = pd.DataFrame(x_train_scaled)\n",
        "\n",
        "x_test_scaled = scaler.transform(x_test)\n",
        "x_test = pd.DataFrame(x_test_scaled)\n",
        "\n",
        "# Define the KNN regression function\n",
        "def knn_regression(k,x_train,y_train,x_test,y_test):\n",
        "    model = neighbors.KNeighborsRegressor(n_neighbors = k)\n",
        "    model.fit(x_train, y_train)\n",
        "    pred=model.predict(x_test)\n",
        "    error = sqrt(mean_squared_error(y_test,pred))\n",
        "    rmse_val.append(error)\n",
        "\n",
        "# Apply KNN regression for different values of k\n",
        "rmse_val = []\n",
        "for k in range(1,21):\n",
        "  knn_regression(k,x_train,y_train,x_test,y_test)\n",
        "\n",
        "# Plot the RMSE values\n",
        "figure = plt.figure(figsize=(10,10))\n",
        "plt.plot(np.arange(1,21),rmse_val)\n",
        "plt.xlabel(\"K\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.show()\n",
        "\n",
        "# Print the minimum error and the corresponding k value\n",
        "rmse_val = np.asarray(rmse_val)\n",
        "print(\"Minimum error {} is at k = {}\".format(np.min(rmse_val),np.argmin(rmse_val)+1))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 871
        },
        "id": "-mcLAkyAymLM",
        "outputId": "d23a14d5-e7b0-4097-a28f-f27c60f459a4"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x1000 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Minimum error 0.14312970031098057 is at k = 8\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "I performed the knn regression on sklearn's iris dataset and observed that minimum error was found at k = 8"
      ],
      "metadata": {
        "id": "ylNj3T4r0eQm"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "---\n",
        "<h1>End of the lab\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "e3wgGwF80v2j"
      }
    }
  ]
}