3. KNN.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "id": "4961daf7-8e1a-4e76-baac-655f419b8f6a",
   "metadata": {},
   "source": [
    "# K-Nearest Neighbour (KNN)\n",
    "The K-Nearest Neighbors (KNN) algorithm is a popular machine learning technique used for classification and regression tasks. It relies on the idea that similar data points tend to have similar labels or values.<br>\n",
    "During the training phase, the KNN algorithm stores the entire training dataset as a reference. When making predictions, it calculates the distance between the input data point and all the training examples, using a chosen distance metric such as Euclidean distance, Manhattan distance, etc according to the dataset.<br>\n",
    "Next, the algorithm identifies the K nearest neighbors to the input data point based on their distances. In the case of classification, the algorithm assigns the most common class label among the K neighbors as the predicted label for the input data point. For regression, it calculates the average or weighted average of the target values of the K neighbors to predict the value for the input data point.<br>\n",
    "The KNN algorithm is straightforward and easy to understand, making it a popular choice in various domains. However, its performance can be affected by the choice of K and the distance metric, so careful parameter tuning is necessary for optimal results.<br>\n",
    "\n",
    "## When to use KNN?\n",
    "KNN Algorithm can be used for both classification and regression predictive problems. However, it is more widely used in classification problems in the industry. To evaluate any technique, we generally look at 3 important aspects:<br>\n",
    "1. Ease of interpreting output\n",
    "2. Calculation time\n",
    "3. Predictive power\n",
    "\n",
    "## How does the KNN Algo Works?\n",
    "Given below is a spread of **red circles (RC)** and **green squares (GS)**:<br>\n",
    "<img src=\"Images/KNN1.png\" width=\"300\" height=\"100\"><br>\n",
    "We intend to find out the class of the **blue star (BS)**. BS can either be RC or GS and nothing else. The **\"K\"** in KNN algorithm is the nearest neighbor we wish to take the vote from. Let’s say **K = 3**. Hence, we will now make a circle with BS as the center just as big as to enclose only three data points on the plane. Refer to the following diagram for more details:<br>\n",
    "<img src=\"Images/KNN2.png\" width=\"200\" height=\"50\"><br>\n",
    "The three closest points to BS are all RC. Hence, with a good confidence level, we can say that the BS should belong to the class RC. Here, the choice became obvious as all three votes from the closest neighbor went to RC. The choice of the parameter K is very crucial in this algorithm.<br>\n",
    "\n",
    "## How to Choose the Factor K?\n",
    "First, let us try to understand the influence of the K-nearest neighbors (KNN) in the algorithm. If we consider the last example, keeping all 6 training observations constant, a given K value allows us to establish **boundaries** for each class. These decision boundaries effectively segregate, for instance, RC from GS. Similarly, let’s examine the impact of the value \"K\" on these class boundaries. The following illustrates the distinct boundaries that separate the two classes, each corresponding to different values of K.<br>\n",
    "<img src=\"Images/KNN3.png\" width=\"400\" height=\"150\"><br>\n",
    "We can see that the boundary becomes **smoother** with **increasing** value of K. With K increasing to **infinity** it finally becomes all blue or all red depending on the total majority.  The training error rate and the validation error rate are two parameters we need to access different K-value. Following is the curve for the training error rate with a varying value of K:<br>\n",
    "<img src=\"Images/KNN4.png\" width=\"400\" height=\"150\"><br>\n",
    "As we can see, the error rate at **K=1** is always **zero** for the training sample. This is because the **closest point** to any training data point is **itself**.Hence the prediction is always accurate with K=1. If validation error curve would have been similar, our choice of K would have been 1. Following is the validation error curve with varying value of K:<br>\n",
    "<img src=\"Images/KNN5.png\" width=\"400\" height=\"150\"><br>\n",
    "At K=1, we were **overfitting** the boundaries. Hence, error rate initially decreases and reaches a minima. After the minima point, it then increase with increasing K. To get the optimal value of K, we can segregate the training and validation from the initial dataset. Now plot the validation error curve to get the optimal value of K. This value of K should be used for all predictions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "220bed85",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn import datasets\n",
    "from sklearn.model_selection import train_test_split\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.colors import ListedColormap\n",
    "cmap = ListedColormap(['#FF0000', '#00FF00', '#0000FF'])\n",
    "from collections import Counter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b694d18b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "iris = datasets.load_iris()\n",
    "X,y = iris.data, iris.target\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 1234)\n",
    "\n",
    "plt.figure()\n",
    "plt.scatter(X[:,2], X[:,3], c=y, cmap=cmap, edgecolor='k', s=20)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e5bd794a-f8f3-4597-8fe1-80f70fae083a",
   "metadata": {},
   "source": [
    "We are loading the iris dataset using sklearn library **`datasets.load_iris()`** this dataset contains features of iris flowers and their corresponding labels.<br>\n",
    "Then we are splitting the dataset into train and test set using **`train_test_split()`** function from sklearn library.<br>\n",
    "To view the data we will create a scatter plot of train set using **`plt.scatter()`**, the **cmap** parameter specifies the colours to use for points. The **edgecolor='k'** parameter sets the colour of the point edges to black and **s=20** specifies the size of the points."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a74a1634",
   "metadata": {},
   "outputs": [],
   "source": [
    "def euclidean_distance(x1,x2):\n",
    "    distances = np.sqrt(np.sum((x1-x2)**2))\n",
    "    return distances"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3a2b6e2-dba9-4852-9170-8c10c8383444",
   "metadata": {},
   "source": [
    "**`euclidean_distance()`** function calculates the Euclidean distance between two vectors **x1** and **x2**. The Euclidean distance is a measure of the straight-line distance between two points in Euclidean space."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "efcf0b68-3856-4e45-a020-5cbc80c7fba4",
   "metadata": {},
   "source": [
    "# KNN for Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a381cb36",
   "metadata": {},
   "outputs": [],
   "source": [
    "class KNN:\n",
    "    def __init__(self, k=3):\n",
    "        self.k = k\n",
    "        \n",
    "    def fit(self, X, y):\n",
    "        self.X_train = X\n",
    "        self.y_train = y\n",
    "        \n",
    "    def predict(self, X):\n",
    "        predictions = [self._predict(x) for x in X]\n",
    "        return predictions\n",
    "    \n",
    "    def _predict(self, x):\n",
    "        # compute the distance\n",
    "        distances = [euclidean_distance(x, x_train) for x_train in self.X_train]\n",
    "        \n",
    "        # get the closest k\n",
    "        k_indices = np.argsort(distances)[:self.k]\n",
    "        k_nearest_labels = [self.y_train[i] for i in k_indices]\n",
    "        \n",
    "        # majority vote\n",
    "        most_common = Counter(k_nearest_labels).most_common()\n",
    "        return most_common[0][0]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a9e2587a-6b88-4142-b61e-7df3c106f9cf",
   "metadata": {},
   "source": [
    "The above class implements the KNN algorithm:<br>\n",
    "The **`__init__()`** is a constructor method that initializes the KNN object with a parameter k, which specifies the number of nearest neighbors to consider when making predictions. By default, k is set to 3.<br>\n",
    "The **`fit()`** method is used to train the KNN model. It takes two parameters **X** and **y**, where X is the training data (a 2D array-like structure where each row represents a sample and each column represents a feature) and y is the corresponding target labels for each sample in X. The method stores X and y in the **self.X_train** and **self.y_train** attributes, respectively.<br>\n",
    "The **`predict()`** method is used to make predictions on new data. It takes a 2D array-like structure X as input, where each row represents a new sample for which predictions need to be made. It iterates over each sample in X and calls the **_predict** method to make individual predictions for each sample. It returns a **list of predictions** for all the samples in X.<br>\n",
    "The **`_predict()`** is a **private** method that is used to make a prediction for a single sample x. It computes the **Euclidean distance** between x and each sample in the training data **self.X_train** to find the k nearest neighbors. It then performs a majority vote among the labels of the k nearest neighbors to determine the predicted label for x."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a5e2a467",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 2, 2, 0, 1, 0, 0, 0, 1, 2, 1, 0, 2, 1, 0, 1, 2, 0, 2, 1, 1, 1, 1, 1, 2, 0, 2, 1, 2, 0]\n"
     ]
    }
   ],
   "source": [
    "clf = KNN(k=5)\n",
    "clf.fit(X_train, y_train)\n",
    "preds = clf.predict(X_test)\n",
    "\n",
    "print(preds)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "306a1d67-7a84-4546-a779-8c8d658be248",
   "metadata": {},
   "source": [
    "**`clf = KNN(k=5)`** creates an instance of the KNN class with k=5, which means the model will consider the 5 nearest neighbors for classification.<br>\n",
    "**`clf.fit(X_train, y_train)`** fits the KNN model to the training data X_train and corresponding labels y_train. It stores this training data internally within the KNN object.<br>\n",
    "**`preds = clf.predict(X_test)`** makes predictions on the test data X_test using the trained KNN model. It calls the predict method of the KNN object, which returns a list of predicted labels for each sample in X_test.<br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "788e6d65",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9666666666666667\n"
     ]
    }
   ],
   "source": [
    "acc = np.sum(preds == y_test)/len(y_test)\n",
    "print(acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d68a20e9-26f3-4e6b-b2a8-b663135524c7",
   "metadata": {},
   "source": [
    "**`acc = np.sum(preds == y_test)/len(y_test)`** calculates the accuracy of the predictions by comparing each element of the **preds** array (predicted labels) with the corresponding element in the **y_test** array (actual labels). The expression **preds == y_test** creates a **boolean** array where True indicates a correct prediction and False indicates an incorrect prediction. **np.sum(preds == y_test)** counts the number of correct predictions, and dividing it by the number of the test samples gives the accuracy. So here accuracy is the ratio of correct predictions to the total number of predictions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "14b45d5a-820c-41d0-a775-d0c6b87728df",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.datasets import make_classification, make_regression\n",
    "\n",
    "# Generate a classification dataset\n",
    "X_clf, y_clf = make_classification(n_samples=100, n_features=2, n_informative=2, n_redundant=0, random_state=42)\n",
    "X_c_train, X_c_test, y_c_train, y_c_test = train_test_split(X_clf, y_clf, test_size = 0.2, random_state = 1234)\n",
    "plt.figure()\n",
    "plt.scatter(X_clf[:, 0], X_clf[:, 1], c=y_clf, cmap='viridis')\n",
    "plt.title(\"Classification Dataset\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d9cc454b-5f7c-46d6-9362-e98c17bd9463",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Generate a regression dataset\n",
    "X_reg, y_reg = make_regression(n_samples=100, n_features=1, noise=10, random_state=42)\n",
    "X_r_train, X_r_test, y_r_train, y_r_test = train_test_split(X_reg, y_reg, test_size = 0.2, random_state = 1234)\n",
    "plt.figure()\n",
    "plt.scatter(X_reg, y_reg, color='b')\n",
    "plt.title(\"Regression Dataset\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4b73a8e-d6b5-44d6-94b2-71243619bb4d",
   "metadata": {},
   "source": [
    "Here we have created dataset for both **Regression** and **Classification** tasks and we will perform both this tasks using KNN."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9aa7ca46-5f7d-4333-b8fe-d8a18f01facb",
   "metadata": {},
   "source": [
    "# KNN for Classification and Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d40fe60d-b2fb-43fe-adc4-803083089f2b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "import numpy as np\n",
    "\n",
    "class KNN:\n",
    "    def __init__(self, k=3):\n",
    "        self.k = k\n",
    "        \n",
    "    def fit(self, X, y):\n",
    "        self.X_train = X\n",
    "        self.y_train = y\n",
    "        \n",
    "    def predict(self, X, task='classification'):\n",
    "        if task == 'classification':\n",
    "            return self._predict_classification(X)\n",
    "        elif task == 'regression':\n",
    "            return self._predict_regression(X)\n",
    "        else:\n",
    "            raise ValueError(\"Invalid task. Supported tasks are 'classification' and 'regression'.\")\n",
    "\n",
    "    def _predict_classification(self, X):\n",
    "        predictions = [self._predict_single_classification(x) for x in X]\n",
    "        return np.array(predictions)\n",
    "    \n",
    "    def _predict_single_classification(self, x):\n",
    "        distances = [euclidean_distance(x, x_train) for x_train in self.X_train]\n",
    "        k_indices = np.argsort(distances)[:self.k]\n",
    "        k_nearest_labels = [self.y_train[i] for i in k_indices]\n",
    "        most_common = Counter(k_nearest_labels).most_common(1)\n",
    "        return most_common[0][0]\n",
    "\n",
    "    def _predict_regression(self, X):\n",
    "        predictions = [self._predict_single_regression(x) for x in X]\n",
    "        return np.array(predictions)\n",
    "\n",
    "    def _predict_single_regression(self, x):\n",
    "        distances = [euclidean_distance(x, x_train) for x_train in self.X_train]\n",
    "        k_indices = np.argsort(distances)[:self.k]\n",
    "        k_nearest_labels = [self.y_train[i] for i in k_indices]\n",
    "        return np.mean(k_nearest_labels)\n",
    "\n",
    "def euclidean_distance(x1, x2):\n",
    "    return np.sqrt(np.sum((x1-x2)**2))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71828b54-c79e-499f-9a79-6f19fc64fe1a",
   "metadata": {},
   "source": [
    "In this class the **`_predict_regression()`** function uses the **`_predict_single_regression()`** method to find the distance between the given point with its k-nearest neighbours and take the mean of all the distances. Similarly the **`_predict_classification()`** function uses the **`_predict_single_classification()`** method to find the distance between the given point with its k-nearest neighbours find the closest point and assign the label of this point to the given point. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "38918175-6193-4fa6-a3b6-8c99c8d040b4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 1, 2, 0, 1, 0, 0, 0, 1, 2, 1, 0, 2, 1, 0, 1, 2, 0, 2, 1, 1, 1,\n",
       "       1, 1, 2, 0, 2, 1, 2, 0])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Classification example\n",
    "clf = KNN(k=3)\n",
    "clf.fit(X_train, y_train)\n",
    "y_pred_classification = clf.predict(X_test, task='classification')\n",
    "y_pred_classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "4a3bab82-bcd3-4762-a1f5-93492fe0fe3d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1.        , 1.33333333, 2.        , 0.        , 1.        ,\n",
       "       0.        , 0.        , 0.        , 1.        , 2.        ,\n",
       "       1.        , 0.        , 2.        , 1.        , 0.        ,\n",
       "       1.        , 2.        , 0.        , 2.        , 1.        ,\n",
       "       1.        , 1.        , 1.        , 1.        , 2.        ,\n",
       "       0.        , 2.        , 1.        , 2.        , 0.        ])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Regression example\n",
    "reg = KNN(k=3)\n",
    "reg.fit(X_train, y_train)\n",
    "y_pred_regression = reg.predict(X_test, task='regression')\n",
    "y_pred_regression"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "378021e7-24fa-4757-ad0d-d48c05bfed0b",
   "metadata": {},
   "source": [
    "# Using Sklearn Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "970fe812-6ae1-49c8-bc90-229215807325",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 1.0\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "iris = load_iris()\n",
    "X = iris.data\n",
    "y = iris.target\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "knn_classifier = KNeighborsClassifier(n_neighbors=3)\n",
    "knn_classifier.fit(X_train, y_train)\n",
    "\n",
    "y_pred = knn_classifier.predict(X_test)\n",
    "\n",
    "accuracy = accuracy_score(y_test, y_pred)\n",
    "print(\"Accuracy:\", accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "50a6527f-75e1-42cb-8c66-139862e9273f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Squared Error: 3364.3932584269664\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import load_diabetes\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "diabetes = load_diabetes()\n",
    "X = diabetes.data\n",
    "y = diabetes.target\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "knn_regressor = KNeighborsRegressor(n_neighbors=3)\n",
    "knn_regressor.fit(X_train, y_train)\n",
    "\n",
    "y_pred = knn_regressor.predict(X_test)\n",
    "\n",
    "mse = mean_squared_error(y_test, y_pred)\n",
    "print(\"Mean Squared Error:\", mse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "56a6c080-01f6-4bfe-9311-5618a0ed97aa",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}