{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# <center><div style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\">Complete 9 Dimensionality Reduction Techniques Comparison using the Kannada MNIST Data </div></center>\n","metadata":{}},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 15pt\">Competition Overview</span>**\n\n__Kannada MNIST__\nMNIST like datatset for Kannada handwritten digits\n\nThe goal of this competition is to provide a simple extension to the classic MNIST competition we're all familiar with. Instead of using Arabic numerals, it uses a recently-released dataset of Kannada digits.\n\nKannada is a language spoken predominantly by people of Karnataka in southwestern India. The language has roughly 45 million native speakers and is written using the Kannada script\n\n![kannada.png](attachment:kannada.png)\n\n\nThis competition uses the same format as the MNIST competition in terms of how the data is structured, but it's different in that it is a synchronous re-run Kernels competition. You write your code in a Kaggle Notebook, and when you submit the results, your code is scored on both the public test set, as well as a private (unseen) test set.\nTechnical Information\n\nAll details of the dataset curation has been captured in the paper titled: Prabhu, Vinay Uday. \"Kannada-MNIST: A new handwritten digits dataset for the Kannada language.\" arXiv preprint arXiv:1908.01242 (2019)\n\nThe github repo of the author can be found here.\n\nOn the originally-posted dataset, the author suggests some interesting questions you may be interested in exploring. Please note, although this dataset has been released in full, the purpose of this competition is for practice, not to find the labels to submit a perfect score.\n\nIn addition to the main dataset, the author also disseminated an additional real world handwritten dataset (with 10k images), termed as the 'Dig-MNIST dataset' that can serve as an out-of-domain test dataset. It was created with the help of volunteers that were non-native users of the language, authored on a smaller sheet and scanned with different scanner settings compared to the main dataset. This 'dig-MNIST' dataset serves as a more difficult test-set (An accuracy of 76.1% was reported in the paper cited above) and achieving ~98+% accuracy on this test dataset would be rather commendable\n\n\n__Files:__\n\n* __train.csv__ - the training set\n* __test.csv__ - the test set\n* __sample_submission.csv__ - a sample submission file in the correct format\n* __Dig-MNIST.csv__ - an additional labeled set of characters that can be used to validate or test model results before submitting to the leaderboard\n\n\n__Dataset Description:__\n\n* The data files train.csv and test.csv contain gray-scale images of hand-drawn digits, from zero through nine, in the Kannada script.\n\n* Each image is 28 pixels in height and 28 pixels in width, for a total of 784 pixels in total. Each pixel has a single pixel-value associated with it, indicating the lightness or darkness of that pixel, with higher numbers meaning darker. This pixel-value is an integer between 0 and 255, inclusive.\n\n* The training data set, train.csv, has 785 columns. The first column, called label, is the digit that was drawn by the user. The rest of the columns contain the pixel-values of the associated image.\n\n* Each pixel column in the training set has a name like pixel{x}, where x is an integer between 0 and 783, inclusive. To locate this pixel on the image, suppose that we have decomposed x as x = i * 28 + j, where i and j are integers between 0 and 27, inclusive. Then pixel{x} is located on row i and column j of a 28 x 28 matrix, (indexing by zero).\n\n\n__Important Note :__\n\n* __i will perform 9 dimensionality reduction techniques to reduce pixels columns from 784 pixels columns to 50 using n_components=50 across all dimensionality reduction techniques.__","metadata":{},"attachments":{"kannada.png":{"image/png":"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"}}},{"cell_type":"markdown","source":"## **<center><span style=\"color:white;background-color:#000000;border-radius: 5px;padding: 5px; font-size: 18pt\">If you find this notebook useful or interesting, please, support with an upvote :)</span></center>**","metadata":{}},{"cell_type":"markdown","source":"<a id=\"1\"></a>\n## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 15pt\">Introduction to Dimensionality Reduction:</span>**\n__Dimensionality reduction is a set of techniques in machine learning used to reduce the number of features in a dataset while retaining as much of the important information as possible. In other words, it is a process of transforming high-dimensional data into a lower-dimensional space that still preserves the essence of the original data.__\n\n__In machine learning, high-dimensional data refers to data with a large number of features or variables. The curse of dimensionality is a common problem in machine learning, where the performance of the model deteriorates as the number of features increases. This is because the complexity of the model increases with the number of features, and it becomes more difficult to find a good solution. In addition, high-dimensional data can also lead to overfitting, where the model fits the training data too closely and does not generalize well to new data.__\n\n__Dimensionality reduction can help to mitigate these problems by reducing the complexity of the model and improving its generalization performance This is important because high-dimensional data can be difficult to work with, and can result in slow and computationally expensive models. We will discuss the importance of dimensionality reduction and Advantages/Disadvantages and some of the most commonly used techniques.__\n\n## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 15pt\">Importance of Dimensionality Reduction:</span>**\n\n* __1.Better performance__\n\nOne of the primary reasons to use dimensionality reduction techniques is to improve the performance of machine learning models. High-dimensional data can result in overfitting, where the model learns to fit the training data too closely and performs poorly on new, unseen data. By reducing the number of dimensions in the dataset, we can reduce the complexity of the model and prevent overfitting.\n\n* __2.Reduced computational cost__\n\nAnother important reason to use dimensionality reduction is to reduce the computational cost of working with high-dimensional data. Many machine learning algorithms, such as clustering and classification, become computationally expensive as the number of dimensions in the dataset increases. By reducing the dimensionality of the data, we can speed up these algorithms and make them more efficient.\n\n* __3.Improved interpretability__\n\nHigh-dimensional data can be difficult to interpret and visualize. By reducing the number of dimensions in the dataset, we can make it easier to understand and explore the data. This can lead to better insights and more informed decision-making.\n\n\n## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 15pt\">Advantages of Dimensionality Reduction:</span>**\n* It helps in data compression, and hence reduced storage space.\n* It reduces computation time.\n* It also helps remove redundant features, if any.\n* Improved Visualization: High dimensional data is difficult to visualize, and dimensionality reduction techniques can help in visualizing the data in 2D or 3D, which can help in better understanding and analysis.\n* Overfitting Prevention: High dimensional data may lead to overfitting in machine learning models, which can lead to poor generalization performance. Dimensionality reduction can help in reducing the complexity of the data, and hence prevent overfitting.\n* Feature Extraction: Dimensionality reduction can help in extracting important features from high dimensional data, which can be useful in feature selection for machine learning models.\n* Data Preprocessing: Dimensionality reduction can be used as a preprocessing step before applying machine learning algorithms to reduce the dimensionality of the data and hence improve the performance of the model.\n* Improved Performance: Dimensionality reduction can help in improving the performance of machine learning models by reducing the complexity of the data, and hence reducing the noise and irrelevant information in the data.\n\n## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 15pt\">Disadvantages of Dimensionality Reduction:</span>**\n\n* It may lead to some amount of data loss.\n* PCA tends to find linear correlations between variables, which is sometimes undesirable.\n* PCA fails in cases where mean and covariance are not enough to define datasets.\n* We may not know how many principal components to keep- in practice, some thumb rules are applied.\n* Interpretability: The reduced dimensions may not be easily interpretable, and it may be difficult to understand the relationship between the original features and the reduced dimensions.\n* Overfitting: In some cases, dimensionality reduction may lead to overfitting, especially when the number of components is chosen based on the training data.\n* Sensitivity to outliers: Some dimensionality reduction techniques are sensitive to outliers, which can result in a biased representation of the data.\n* Computational complexity: Some dimensionality reduction techniques, such as manifold learning, can be computationally intensive, especially when dealing with large datasets.\n\n## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 15pt\">Important Points:</span>**\n\n* Dimensionality reduction is the process of reducing the number of features in a dataset while retaining as much information as possible.\n* This can be done to reduce the complexity of a model, improve the performance of a learning algorithm, or make it easier to visualize the data.\n* Each technique projects the data onto a lower-dimensional space while preserving important information.\n* Dimensionality reduction is performed during pre-processing stage before building a model to improve the performance\n* It is important to note that dimensionality reduction can also discard useful information, so care must be taken when applying these techniques.\n\n\n## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 15pt\">Dimensionality Reduction Techniques:</span>**\n\n* __1.PCA (Principal Component Analysis)__\n* __2.FastICA (Independent Component Analysis)__\n* __3.UMAP (Uniform Manifold Approximation and Projection)__\n* __4.TruncatedSVD (Truncated Singular Value Decomposition)__\n* __5.Factor Analysis__\n* __6.t-SNE (t-Distributed Stochastic Neighbor Embedding)__\n* __7.NMF (Non-negative Matrix Factorization)__\n* __8.Gaussian Random Projection__\n* __9.Sparse Random Projection__","metadata":{}},{"cell_type":"markdown","source":"<a id=\"top\"></a>\n<div class=\"list-group\" id=\"list-tab\" role=\"tablist\">\n    \n  <h3 class=\"list-group-item list-group-item-action active\" data-toggle=\"list\" role=\"tab\" aria-controls=\"home\" style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; line-height: 1;\">Table of Contents</h3>\n\n</div>\n\n\n* [1. Introduction](#1)\n\n* [2. Importing Libraries](#2)    \n\n* [3. Data Preprocessing](#3)\n    \n* [4. Defining Helper Functions](#4)\n    \n* [5. Exploring the Impact of Dimensionality Reduction Techniques on KNeighborsClassifier Accuracy](#5)\n    \n    * [5.1. PCA (Principal Component Analysis)](#5.1)\n    * [5.2. FastICA (Independent Component Analysis)](#5.2)\n    * [5.3. UMAP (Uniform Manifold Approximation and Projection)](#5.3)\n    * [5.4. TruncatedSVD (Truncated Singular Value Decomposition)](#5.4)\n    * [5.5. FactorAnalysis](#5.5)\n    * [5.6. t-SNE (t-distributed Stochastic Neighbor Embedding)](#5.6)\n    * [5.7. NMF (Non-Negative Matrix Factorization)](#5.7)\n    * [5.8. GaussianRandomProjection](#5.8)\n    * [5.9. SparseRandomProjection](#5.9)\n\n* [6. Final Results](#6) \n* [7. Best Model and Competition Submission](#7) \n\n","metadata":{}},{"cell_type":"markdown","source":"<a id=\"2\"></a>\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> :::    Import Libraries ::: </b></font>\n\n---\n<a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41;\" data-toggle=\"popover\">Go to Table of Contents</a>","metadata":{}},{"cell_type":"code","source":"import warnings\nwarnings.filterwarnings('ignore')\nimport numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport seaborn as sns\nfrom sklearn.decomposition import PCA, FastICA, NMF, TruncatedSVD, FactorAnalysis\nfrom sklearn.manifold import TSNE\nfrom umap import umap_ as umap\nfrom sklearn.random_projection import GaussianRandomProjection, SparseRandomProjection\nfrom sklearn.preprocessing import StandardScaler\nfrom sklearn.pipeline import Pipeline\nfrom sklearn.neighbors import KNeighborsClassifier\nfrom sklearn.metrics import multilabel_confusion_matrix, confusion_matrix, accuracy_score, precision_score, recall_score, f1_score\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.model_selection import RandomizedSearchCV\nfrom scipy.stats import randint\nimport time\n%matplotlib inline\npd.set_option('display.max_columns', None)","metadata":{"_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"%%html\n<style>\n.output_wrapper .output {\n  overflow-y: visible;\n  height: fit-content;\n}\n</style>\n\nfrom IPython.display import display, HTML\ndisplay(HTML(\"<style>.container { width:100% !important; }</style>\"))\ndisplay(HTML(\"<style>.output_result { max-width:100% !important; }</style>\"))\ndisplay(HTML(\"<style>.prompt { display:none !important; }</style>\"))","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"3\"></a>\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> :::    Data Preprocessing ::: </b></font>\n\n---\n<a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41;\" data-toggle=\"popover\">Go to Table of Contents</a>","metadata":{}},{"cell_type":"code","source":"# Loading the Kannada MNIST Data\ntrain_df = pd.read_csv('/kaggle/input/Kannada-MNIST/train.csv')\nval_df = pd.read_csv('/kaggle/input/Kannada-MNIST/Dig-MNIST.csv')\nX_sample_submission_df = pd.read_csv('/kaggle/input/Kannada-MNIST/test.csv')\ny_sample_submission_df = pd.read_csv('/kaggle/input/Kannada-MNIST/sample_submission.csv')","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df.iloc[:,:10].head()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"val_df.iloc[:,:10].head()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"X_sample_submission_df.iloc[:,:10].head()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_sample_submission_df.head()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#371f41;border-radius: 5px;padding: 5px; font-size: 15pt\"> Separate the features and target labels </span>**","metadata":{}},{"cell_type":"code","source":"X = train_df.iloc[:, 1:].values\ny = train_df.iloc[:, 0].values\nX_val = val_df.iloc[:, 1:].values\ny_val = val_df.iloc[:, 0].values\nX_submission_test = X_sample_submission_df.iloc[:, 1:].values","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(f\"X Shape: {X.shape}.\\ny Shape: {y.shape}.\")\nprint(f\"X_val Shape: {X_val.shape}.\\ny_val Shape: {y_val.shape}.\")\nprint(f\"X_submission_test Shape: {X_submission_test.shape}.\")","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#371f41;border-radius: 5px;padding: 5px; font-size: 15pt\"> Splitting the data </span>**","metadata":{}},{"cell_type":"code","source":"X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(f\"X_train Shape: {X_train.shape}.\\ny_train Shape: {y_train.shape}.\")\nprint(f\"X_test Shape: {X_test.shape}.\\ny_test Shape: {y_test.shape}.\")\nprint(f\"X_val Shape: {X_val.shape}.\\ny_val Shape: {y_val.shape}.\")\nprint(f\"X_submission_test Shape: {X_submission_test.shape}.\")","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#371f41;border-radius: 5px;padding: 5px; font-size: 15pt\"> Scaling the data </span>**","metadata":{}},{"cell_type":"code","source":"scaler = StandardScaler()\nX_train_scaled = scaler.fit_transform(X_train)\nX_test_scaled = scaler.transform(X_test)\nX_val_scaled = scaler.transform(X_val)\nX_submission_test_scaled = scaler.transform(X_submission_test)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### **<span style=\"color:white;background-color:#371f41;border-radius: 5px;padding: 5px; font-size: 15pt\">Plotting Actual Digits Images</span>**","metadata":{}},{"cell_type":"code","source":"plt.figure(figsize=(20,10))\nfor i in range(40):  \n    plt.subplot(4, 10, i+1)\n    plt.imshow(X_train[i].reshape((28,28)),cmap=plt.cm.binary)\n    plt.title(\"Actual=%d\" % y_train[i],y=0.9)\n    plt.axis('off')\nplt.subplots_adjust(wspace=0.3, hspace=-0.1)\nplt.show()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"4\"></a>\n\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> :::    Defining Helper Functions ::: </b></font>\n\n---\n<a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n\n* #### Creating  base dataframe to add all results from all Dimensionality Reduction Techniques and KNN combinations\n* #### Creating  time dataframe to add elapsed time for each Dimensionality Reduction Technique and KNN implementation","metadata":{}},{"cell_type":"code","source":"test_results_df = pd.DataFrame(columns=['Combination Name', 'Test Accuracy', 'Validation Accuracy','knn__n_neighbors',\n                                   'knn__weights','knn__algorithm','knn__leaf_size','knn__p','knn__metric'])\n\ntime_df = pd.DataFrame(columns=[\"DC_Time\",\"KNN_DC_Time\"])","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 15pt\"> KNeighborsClassifier Model Function </span>**\n\n### Please note that i will use cv=5 and n_iter=1 due to time consuming issues :) so feel free to increase and edit any hyperparameters","metadata":{}},{"cell_type":"code","source":"# Define the function\ndef apply_KNN_dc(X_train_dc, X_test_dc, X_val_dc, combination_name=\"KNN_PCA\"):\n    test_results_df = pd.DataFrame(columns=['Combination Name', 'Test Accuracy', 'Validation Accuracy','knn__n_neighbors',\n                                   'knn__weights','knn__algorithm','knn__leaf_size','knn__p','knn__metric'])\n    # Define the pipeline\n    knn_pipeline = Pipeline([\n        ('knn', KNeighborsClassifier())\n    ])\n\n    # Define the hyperparameters to tune\n    knn_params = {\n        'knn__n_neighbors': np.arange(1, 21),\n        'knn__weights': ['uniform', 'distance'],\n        'knn__algorithm': ['auto', 'ball_tree', 'kd_tree', 'brute'],\n        'knn__leaf_size': np.arange(10, 51),\n        'knn__p': [1, 2],\n        'knn__metric': ['euclidean', 'manhattan', 'minkowski']\n    }\n\n    # Fit the pipeline with randomized search CV\n    knn_random_search = RandomizedSearchCV(knn_pipeline, knn_params, cv=5, n_jobs=-1, n_iter=1, verbose=1, random_state=42)\n    knn_random_search.fit(X_train_dc, y_train)\n\n    # Get the best estimator from the randomized search\n    best_knn = knn_random_search.best_estimator_\n    best_params = knn_random_search.best_params_\n    \n    # Evaluate the model on the test set\n    y_test_pred = best_knn.predict(X_test_dc)\n    test_accuracy = accuracy_score(y_test, y_test_pred)\n\n    # Evaluate the model on the validation set\n    y_val_pred = best_knn.predict(X_val_dc)\n    val_accuracy = accuracy_score(y_val, y_val_pred)\n    \n    # Getting method name, scores, best params and save to cn_info dict\n    dc_results = {'Combination Name':combination_name, 'Test Accuracy':test_accuracy, 'Validation Accuracy':val_accuracy}\n    dc_results.update(best_params)\n    \n    # Get the Multilabel confusion matrix\n    labels = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9] # order is important\n    mcm = multilabel_confusion_matrix(y_test, y_test_pred, labels=labels)\n    \n    # Print best SVC Hyperparameters\n    print(f\"Best {combination_name} Hyperparameters: {best_params}\" )\n\n    # Print the evaluation metrics\n    print(f\"{combination_name} Test Accuracy: {test_accuracy}\")\n    print(f\"{combination_name} Validation Accuracy: {val_accuracy}\")\n    \n    fig, ax = plt.subplots(4, 3, figsize=(20, 30))\n    \n    for axes, cfs_matrix, label in zip(ax.flatten(), mcm, labels):\n        print_confusion_matrix(cfs_matrix, axes, label, index, columns)\n        ax[3,1].set_axis_off()\n        ax[3,2].set_axis_off()\n        fig.tight_layout()\n\n    fig.suptitle(combination_name + \" Multilabel Confusion Matrices\", color='#000000', fontsize=20, fontweight='bold')\n    fig.subplots_adjust(top=0.95)\n    plt.show()\n    \n    return dc_results","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 15pt\"> Ploting Multilabel Confusion Matrices Function </span>**","metadata":{}},{"cell_type":"code","source":"labels = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9] # order is important\nindex= ['Actual Positive', 'Actual Negative']\ncolumns= ['Predicted Positive', 'Predicted Negative']\ndef print_confusion_matrix(confusion_matrix, axes, class_label, ind , col, fontsize=14):\n\n    df_cm = pd.DataFrame(\n        confusion_matrix, index=index, columns=columns,\n    )\n\n    try:\n        heatmap = sns.heatmap(df_cm, annot=True, fmt=\"d\",cmap='rocket', cbar=False, square=True,\n                              annot_kws={\"fontsize\":12,'fontweight':'bold'}, ax=axes)\n    except ValueError:\n        raise ValueError(\"Confusion matrix values must be integers.\")\n    heatmap.yaxis.set_ticklabels(heatmap.yaxis.get_ticklabels(), rotation=90, fontsize=14,fontweight='bold')\n    heatmap.xaxis.set_ticklabels(heatmap.xaxis.get_ticklabels(), rotation=0, fontsize=14,fontweight='bold')\n#     axes.set_ylabel('True label')\n#     axes.set_xlabel('Predicted label')\n    axes.set_title(\"Confusion Matrix for digit - \" + str(class_label), color='#000000', fontsize=16, fontweight='bold')","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"5\"></a>\n\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> :::    Exploring the Impact of Dimensionality Reduction Techniques on KNeighborsClassifier Accuracy ::: </b></font>\n\n---\n<a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41;\" data-toggle=\"popover\">Go to Table of Contents</a>\n","metadata":{}},{"cell_type":"markdown","source":"<a id=\"5.1\"></a>\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> ::: PCA (Principal Component Analysis) ::: </b></font> <a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41; margin-left: 35px;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n------------\n\n__PCA is a widely used dimensionality reduction technique that can be used to reduce the number of features in a dataset while retaining as much information as possible. PCA works by finding the directions of maximum variance in a dataset and projecting the data onto these directions. The resulting projection is a lower-dimensional representation of the original data that retains the most important information.__\n\n\n\n__PCA steps:__\n\n* Center the data by subtracting the mean from each feature.\n\n* Compute the covariance matrix of the data. This matrix measures the linear relationship between pairs of features.\n\n* Compute the eigenvectors and eigenvalues of the covariance matrix. The eigenvectors represent the directions in the data with the most variance, while the eigenvalues represent the amount of variance along each direction.\n\n* Sort the eigenvectors by their corresponding eigenvalues in descending order. This ensures that the most important directions are selected first.\n\n* Select the top k eigenvectors to form a new basis for the data. These eigenvectors are called principal components.\n\n* Project the data onto the new basis. This involves taking the dot product of the data with each principal component to obtain a new set of uncorrelated variables.\n\n\n__Pros:__\n* Simple and widely used method for dimensionality reduction.\n* Preserves the global structure of the data.\n* Can be used for data visualization.\n\n__Cons:__\n* Assumes that the data is linearly related.\n* May not work well for nonlinear data.\n* May not perform well if outliers are present in the data\n","metadata":{}},{"cell_type":"code","source":"start_time = time.time()\n\n# Applying PCA\npca = PCA(n_components=50, random_state=42)\nX_train_pca = pca.fit_transform(X_train_scaled)\nX_test_pca = pca.transform(X_test_scaled)\nX_val_pca = pca.transform(X_val_scaled)\nX_submission_test_pca = pca.transform(X_submission_test_scaled)\n\nend_time = time.time()\npca_dc_elapsed_time = end_time - start_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"start_time = time.time()\n\n# Running KNeighborsClassifier on PCA Data\npca_dc_results = apply_KNN_dc(X_train_pca,X_test_pca,X_val_pca,\"KNN_PCA\")\n\nend_time = time.time()\nknn_pca_dc_elapsed_time = end_time - start_time\n\n# Adding elapsed time to time_df\ntime_df.loc[0, 'DC_Time'] = pca_dc_elapsed_time\ntime_df.loc[0, 'KNN_DC_Time'] = knn_pca_dc_elapsed_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"5.2\"></a>\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> ::: FastICA (Independent Component Analysis) ::: </b></font> <a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41; margin-left: 35px;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n------------\n\n\n__FastICA is a nonlinear dimensionality reduction technique that aims to find a low-dimensional representation of high-dimensional data while maximizing the statistical independence of the resulting components. FastICA separates a multivariate signal into independent, non-Gaussian componentsIt works by finding a linear projection of the data that maximizes the non-Gaussianity of the projected components.__\n\n\n__FastICA steps :__\n\n* Center the data by subtracting the mean from each feature.\n\n* Whiten the data by dividing each feature by its standard deviation. This transforms the data into a new space where the features are uncorrelated and have unit variance.\n\n* Initialize a random weight vector for each independent component. These weight vectors are used to linearly combine the whitened features and obtain the independent components.\n\n* Update the weight vectors using a fixed-point iteration until convergence. This involves maximizing a measure of non-Gaussianity for each weight vector.\n\n* Project the data onto the independent components. This involves taking the dot product of the whitened data with each weight vector to obtain a new set of independent components.\n\n\n__Pros:__\n* Can capture nonlinear relationships in the data.\n* Can be used for blind source separation and feature extraction.\n\n__Cons:__\n* May not work well if the data is already close to being statistically independent.\n* Requires careful tuning of parameters.\n* May be sensitive to outliers in the data.\n\n","metadata":{}},{"cell_type":"code","source":"start_time = time.time()\n\n# Applying FastICA\nica = FastICA(n_components=50, random_state=42)\nX_train_ica = ica.fit_transform(X_train_scaled)\nX_test_ica = ica.transform(X_test_scaled)\nX_val_ica = ica.transform(X_val_scaled)\nX_submission_test_ica = ica.transform(X_submission_test_scaled)\n\nend_time = time.time()\nica_dc_elapsed_time = end_time - start_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"start_time = time.time()\n\n# Running KNeighborsClassifier on FastICA Data\nica_dc_results = apply_KNN_dc(X_train_ica,X_test_ica,X_val_ica,\"KNN_FastICA\")\n\n\nend_time = time.time()\nknn_ica_dc_elapsed_time = end_time - start_time\n\n# Adding elapsed time to time_df\ntime_df.loc[1, 'DC_Time'] = ica_dc_elapsed_time\ntime_df.loc[1, 'KNN_DC_Time'] = knn_ica_dc_elapsed_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"5.3\"></a>\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> ::: UMAP (Uniform Manifold Approximation and Projection) ::: </b></font> <a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41; margin-left: 35px;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n------------\n\nUMAP is a nonlinear dimensionality reduction technique that aims to find a low-dimensional representation of high-dimensional data while preserving the local and global structure of the data. It works by constructing a graph representation of the data and optimizing a low-dimensional embedding that preserves the graph structure.\n\n__UMAP steps:__\n* Compute the pairwise distances between the data points. This measures the similarity between each pair of data points.\n\n* Construct a fuzzy simplicial set from the pairwise distances. This involves constructing a graph where the nodes represent the data points and the edges represent the pairwise similarities.\n\n* Optimize the embedding using stochastic gradient descent. This involves minimizing a cost function that measures the difference between the high-dimensional distances and the low-dimensional distances.\n\n* Project the data onto the embedding. This involves mapping the high-dimensional data onto the low-dimensional embedding using the optimized transformation.\n\n\n__Pros:__\n* Can capture nonlinear relationships in the data.\n* Can preserve both local and global structure of the data.\n* Can handle both dense and sparse data.\n\n__Cons:__\n* May require careful tuning of parameters.\n* May be sensitive to the choice of distance metric.\n* Can be computationally expensive for large datasets","metadata":{}},{"cell_type":"code","source":"start_time = time.time()\n\n# Apply UMAP\nreducer = umap.UMAP(n_components=50,n_neighbors=30, random_state=42)\nX_train_umap = reducer.fit_transform(X_train_scaled)\nX_test_umap = reducer.transform(X_test_scaled)\nX_val_umap = reducer.transform(X_val_scaled)\nX_submission_test_umap = reducer.transform(X_submission_test_scaled)\n\nend_time = time.time()\numap_dc_elapsed_time = end_time - start_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"start_time = time.time()\n\n# Running KNeighborsClassifier on UMAP Data\numap_dc_results = apply_KNN_dc(X_train_umap,X_test_umap,X_val_umap,\"KNN_UMAP\")\n\n\nend_time = time.time()\nknn_umap_dc_elapsed_time = end_time - start_time\n\n# Adding elapsed time to time_df\ntime_df.loc[2, 'DC_Time'] = umap_dc_elapsed_time\ntime_df.loc[2, 'KNN_DC_Time'] = knn_umap_dc_elapsed_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"5.4\"></a>\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> ::: TruncatedSVD (Truncated Singular Value Decomposition) ::: </b></font> <a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41; margin-left: 35px;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n------------\n\n\n__TruncatedSVD is a linear technique that factorizes a matrix into three matrices, two orthogonal matrices and a diagonal matrix with the singular values.__\n\n\n__TruncatedSVD steps:__\n\n* Center the data by subtracting the mean from each feature.\n\n* Compute the singular value decomposition (SVD) of the data. This involves decomposing the data matrix into three matrices: U, Σ, and V, where U and V are orthogonal matrices and Σ is a diagonal matrix with the singular values.\n\n* Truncate the SVD by selecting the top k singular values and their corresponding left and right singular vectors. This reduces the dimensionality of the data while retaining as much of the relevant information as possible.\n\n* Project the data onto the truncated SVD. This involves taking the dot product of the original data with the truncated matrices to obtain a new set of variables.\n\n\n__Pros:__\n\n* Can be used for sparse data matrices.\n* Requires less memory than full SVD.\n* Can be used for data visualization.\n\n__Cons:__\n\n* Assumes that the data is linearly related.\n* May not work well for nonlinear data.\n* May not perform well if outliers are present in the data.\n","metadata":{}},{"cell_type":"code","source":"start_time = time.time()\n\n# Applying TruncatedSVD\nsvd = TruncatedSVD(n_components=50, random_state=42)\nX_train_svd = svd.fit_transform(X_train_scaled)\nX_test_svd = svd.transform(X_test_scaled)\nX_val_svd = svd.transform(X_val_scaled)\nX_submission_test_svd = svd.transform(X_submission_test_scaled)\n\nend_time = time.time()\nsvd_dc_elapsed_time = end_time - start_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"start_time = time.time()\n\n# Running KNeighborsClassifier on TruncatedSVD Data\nsvd_dc_results = apply_KNN_dc(X_train_svd,X_test_svd,X_val_svd,\"KNN_TruncatedSVD\")\n\n\nend_time = time.time()\nknn_svd_dc_elapsed_time = end_time - start_time\n\n# Adding elapsed time to time_df\ntime_df.loc[3, 'DC_Time'] = svd_dc_elapsed_time\ntime_df.loc[3, 'KNN_DC_Time'] = knn_svd_dc_elapsed_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"5.5\"></a>\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> ::: FactorAnalysis ::: </b></font> <a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41; margin-left: 35px;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n------------\n\n__Factor Analysis is a dimensionality reduction technique that is used to identify latent variables or factors that explain the variance in a set of observed variables. It assumes that each observed variable is a linear combination of a smaller number of unobserved variables or factors. These factors are typically uncorrelated and are used to explain the underlying structure of the data.__\n\n\n__FactorAnalysis steps:__\n* Center the data by subtracting the mean from each feature.\n\n* Initialize a set of factors and a set of loadings. The factors are unobserved variables that represent the underlying structure of the data, while the loadings are the coefficients that relate the factors to the observed variables.\n\n* Update the factors and loadings using an iterative algorithm until convergence. This involves maximizing the likelihood of the observed data given the factors and loadings.\n\n* Compute the scores for each factor, which represent the degree to which each observation is associated with each factor.\n* Project the data onto the factors. This involves taking the dot product of the original data with the loadings to obtain a new set of factor scores.\n\n__Pros:__\n* Can model the correlation structure among the variables.\n* Can be used for data visualization.\n* Can handle missing data.\n\n__Cons:__\n* Assumes that the factors are uncorrelated with each other.\n* May not work well for highly skewed or heavy-tailed data.\n* Requires careful selection of the number of factors.\n","metadata":{}},{"cell_type":"code","source":"start_time = time.time()\n\n# Applying FactorAnalysis\nfa = FactorAnalysis(n_components=50, random_state=42)\nX_train_fa = fa.fit_transform(X_train_scaled)\nX_test_fa = fa.transform(X_test_scaled)\nX_val_fa = fa.transform(X_val_scaled)\nX_submission_test_fa = fa.transform(X_submission_test_scaled)\n\n\nend_time = time.time()\nfa_dc_elapsed_time = end_time - start_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"start_time = time.time()\n\n# Running KNeighborsClassifier on FactorAnalysis Data\nfa_dc_results = apply_KNN_dc(X_train_fa,X_test_fa,X_val_fa,\"KNN_FactorAnalysis\")\n\nend_time = time.time()\nknn_fa_dc_elapsed_time = end_time - start_time\n\n# Adding elapsed time to time_df\ntime_df.loc[4, 'DC_Time'] = fa_dc_elapsed_time\ntime_df.loc[4, 'KNN_DC_Time'] = knn_fa_dc_elapsed_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"5.6\"></a>\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> ::: t-SNE (t-distributed Stochastic Neighbor Embedding) ::: </b></font> <a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41; margin-left: 35px;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n------------\n\n__t-SNE is a nonlinear technique that maps high-dimensional data into a low-dimensional space while preserving the pairwise similarities between the data points.__\n\n__SKlearn :__ t-SNE is a tool to visualize high-dimensional data. It converts similarities between data points to joint probabilities and tries to minimize the Kullback-Leibler divergence between the joint probabilities of the low-dimensional embedding and the high-dimensional data. t-SNE has a cost function that is not convex, i.e. with different initializations we can get different results.\n\n__It is highly recommended to use another dimensionality reduction method (e.g. PCA for dense data or TruncatedSVD for sparse data) to reduce the number of dimensions to a reasonable amount (e.g. 50) if the number of features is very high. This will suppress some noise and speed up the computation of pairwise distances between samples.__\n\n__t-SNE steps :__\n\n* Compute the pairwise similarities between the data points. This involves computing a probability distribution over pairs of data points in the high-dimensional space, with similar points having a higher probability of being selected.\n\n* Construct a probability distribution over the low-dimensional embedding. This involves computing a similar probability distribution over pairs of points in the low-dimensional space.\n\n* Minimize the Kullback-Leibler divergence between the two distributions using gradient descent. This involves iteratively adjusting the position of the data points in the low-dimensional embedding to minimize the difference between the two distributions.\n\n* Project the data onto the embedding. This involves mapping the high-dimensional data onto the low-dimensional embedding using the optimized transformation.\n\n\n__Pros:__\n* Can capture nonlinear relationships in the data.\n* Can preserve both local and global structure of the data.\n* Can be used for data visualization.\n\n__Cons:__\n* May be sensitive to the choice of distance metric.\n* Can be computationally expensive for large datasets.\n* May not be suitable for clustering or classification tasks.\n","metadata":{}},{"cell_type":"code","source":"start_time = time.time()\n\n# Applying TSNE\ntsne = TSNE(n_components=2, random_state=42)\nX_train_tsne = tsne.fit_transform(X_train_pca)\nX_test_tsne = tsne.fit_transform(X_test_pca)\nX_val_tsne = tsne.fit_transform(X_val_pca)\nX_submission_test_tsne = tsne.fit_transform(X_submission_test_pca)\n\nend_time = time.time()\ntsne_dc_elapsed_time = end_time - start_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"start_time = time.time()\n\n# Running KNeighborsClassifier on FactorAnalysis Data\ntsne_dc_results = apply_KNN_dc(X_train_tsne,X_test_tsne,X_val_tsne,\"KNN_TSNE\")\n\nend_time = time.time()\nknn_tsne_dc_elapsed_time = end_time - start_time\n\n# Adding elapsed time to time_df\ntime_df.loc[5, 'DC_Time'] = tsne_dc_elapsed_time\ntime_df.loc[5, 'KNN_DC_Time'] = knn_tsne_dc_elapsed_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"5.7\"></a>\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> ::: NMF (Non-Negative Matrix Factorization) ::: </b></font> <a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41; margin-left: 35px;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n------------\n\n__NMF is a nonlinear dimensionality reduction technique that aims to find a low-dimensional representation of high-dimensional data while constraining the resulting components to be non-negative. It works by factorizing the input matrix into two non-negative matrices, which can be interpreted as basis vectors and coefficients__\n\n__In short __NMF__ is a nonlinear technique that factorizes a non-negative matrix into two non-negative matrices, one with the basis vectors and the other with the coefficients__\n\n__NMF steps:__\n\n* Initialize non-negative matrices W and H. These matrices represent the basis vectors and coefficients, respectively.\n\n* Update the matrices using a multiplicative update rule until convergence. This involves minimizing the Euclidean distance between the original data and the reconstructed data obtained from the product of W and H.\n\n* Project the data onto the matrix W. This involves taking the dot product of the original data with the basis vectors in W to obtain a new set of coefficients.\n\n\n__Pros:__\n\n* Can capture nonlinear relationships in the data.\n* Can be used for feature extraction and clustering.\n* Constraints on non-negativity can lead to more interpretable results.\n\n__Cons:__\n* May be sensitive to the choice of initialization.\n* May not work well for highly sparse or noisy data.\n* Does not preserve the original data scale.\n","metadata":{}},{"cell_type":"code","source":"start_time = time.time()\n\n# Applying NMF\nnmf = NMF(n_components=50, random_state=42)\nX_train_nmf = nmf.fit_transform(X_train)\nX_test_nmf = nmf.transform(X_test)\nX_val_nmf = nmf.transform(X_val)\nX_submission_test_nmf = nmf.transform(X_submission_test)\n\nend_time = time.time()\nnmf_dc_elapsed_time = end_time - start_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"start_time = time.time()\n\n# Running KNeighborsClassifier on NMF Data\nnmf_dc_results = apply_KNN_dc(X_train_nmf,X_test_nmf,X_val_nmf,\"KNN_NMF\")\n\nend_time = time.time()\nknn_nmf_dc_elapsed_time = end_time - start_time\n\n# Adding elapsed time to time_df\ntime_df.loc[6, 'DC_Time'] = nmf_dc_elapsed_time\ntime_df.loc[6, 'KNN_DC_Time'] = knn_nmf_dc_elapsed_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"5.8\"></a>\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> ::: GaussianRandomProjection ::: </b></font> <a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41; margin-left: 35px;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n------------\n\n__GaussianRandomProjection is a dimensionality reduction technique that involves randomly projecting high-dimensional data onto a lower-dimensional space using a matrix with values drawn from a Gaussian distribution. The resulting projected data will have fewer dimensions than the original data, making it easier to analyze and process. GaussianRandomProjection is computationally efficient and can retain important information from the original data, but it may not retain all of the important information and can require tuning of hyperparameters.__\n\n\n__GaussianRandomProjection steps:__\n\n\n* Initialize the number of components to project onto\n* Initialize a random matrix with values drawn from a Gaussian distribution\n* Project the data onto the random matrix by computing the dot product between the data and the matrix\n* Return the projected data, which will have fewer dimensions than the original data\n\n\n\n__Pros:__\n\n* Computationally efficient: GaussianRandomProjection can be faster than other dimensionality reduction techniques like PCA, especially for high-dimensional data.\n* Retains important information: GaussianRandomProjection can retain important information from the original data, making it useful for certain types of machine learning tasks.\n* Easy to implement: GaussianRandomProjection is relatively easy to implement and does not require complex optimization algorithms.\n\n\n__Cons:__\n\n* Loss of information: GaussianRandomProjection may not retain all of the important information from the original data, particularly for datasets with a high degree of correlation.\n* Difficult to choose the number of components: It can be difficult to determine the optimal number of components to project onto, which can affect the quality of the results.\n* Requires tuning: GaussianRandomProjection may require some tuning of hyperparameters to achieve the best results.\n\n","metadata":{}},{"cell_type":"code","source":"start_time = time.time()\n\n# Applying GaussianRandomProjection\ngrp = GaussianRandomProjection(n_components=50, random_state=42)\nX_train_grp = grp.fit_transform(X_train_scaled)\nX_test_grp = grp.transform(X_test_scaled)\nX_val_grp = grp.transform(X_val_scaled)\nX_submission_test_grp = grp.transform(X_submission_test_scaled)\n\nend_time = time.time()\ngrp_dc_elapsed_time = end_time - start_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"start_time = time.time()\n\n# Running KNeighborsClassifier on NMF Data\ngrp_dc_results = apply_KNN_dc(X_train_grp,X_test_grp,X_val_grp,\"KNN_GaussianRandomProjection\")\n\nend_time = time.time()\nknn_grp_dc_elapsed_time = end_time - start_time\n\n# Adding elapsed time to time_df\ntime_df.loc[7, 'DC_Time'] = grp_dc_elapsed_time\ntime_df.loc[7, 'KNN_DC_Time'] = knn_grp_dc_elapsed_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"5.9\"></a>\n### <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> ::: SparseRandomProjection ::: </b></font> <a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41; margin-left: 35px;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n------------\n\n__SparseRandomProjection is another dimensionality reduction technique that involves randomly projecting high-dimensional data onto a lower-dimensional space using a matrix with a specified sparsity level, meaning that a certain percentage of entries in the matrix will be zero. The resulting projected data will have fewer dimensions than the original data, making it easier to analyze and process. SparseRandomProjection is also computationally efficient and can remove noise from the original data, but it may not retain all of the important information and can require tuning of hyperparameters. SparseRandomProjection can produce sparse projections, which can be more interpretable than dense projections, but it may not be suitable for all datasets, particularly datasets with a low degree of sparsity or with a high degree of correlation between features.__\n\n\n__SparseRandomProjection steps :__\n\n\n* Initialize the number of components to project onto\n* Initialize a random matrix with a specified sparsity level, meaning that a certain percentage of entries in the matrix will be zero\n* Project the data onto the random matrix by computing the dot product between the data and the matrix\n* Return the projected data, which will have fewer dimensions than the original data.\n\n\n__Pros:__\n* Computationally efficient: SparseRandomProjection can be faster than other dimensionality reduction techniques, especially for high-dimensional data.\n* Can reduce noise: SparseRandomProjection can remove noise from the original data, which can improve the performance of machine learning algorithms.\n* Can be more interpretable: SparseRandomProjection can produce sparse projections, which can be more interpretable than dense projections.\n\n__Cons:__\n\n* Loss of information: SparseRandomProjection may not retain all of the important information from the original data, particularly for datasets with a high degree of correlation.\n* Requires tuning: SparseRandomProjection may require some tuning of hyperparameters to achieve the best results.\n* May not be suitable for all datasets: SparseRandomProjection may not be suitable for datasets with a low degree of sparsity or for datasets with a high degree of correlation","metadata":{}},{"cell_type":"code","source":"start_time = time.time()\n\n# Applying SparseRandomProjection\nsrp = SparseRandomProjection(n_components=50, random_state=42)\nX_train_srp = srp.fit_transform(X_train_scaled)\nX_test_srp = srp.transform(X_test_scaled)\nX_val_srp = srp.transform(X_val_scaled)\nX_submission_test_srp = srp.transform(X_submission_test_scaled)\n\nend_time = time.time()\nsrp_dc_elapsed_time = end_time - start_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"start_time = time.time()\n\n# Running KNeighborsClassifier on SparseRandomProjection Data\nsrp_dc_results = apply_KNN_dc(X_train_srp,X_test_srp,X_val_srp,\"KNN_SparseRandomProjection\")\n\nend_time = time.time()\nknn_srp_dc_elapsed_time = end_time - start_time\n\n# Adding elapsed time to time_df\ntime_df.loc[8, 'DC_Time'] = srp_dc_elapsed_time\ntime_df.loc[8, 'KNN_DC_Time'] = knn_srp_dc_elapsed_time","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"6\"></a>\n## <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> ::: Final Results ::: </b></font> <a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41; margin-left: 30px;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n___","metadata":{}},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 12pt\">Elapsed time for each Dimensionality Reduction Technique and KNN implementation : </span>**","metadata":{}},{"cell_type":"code","source":"time_df","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 12pt\">Combining all results of all Dimensionality Reduction Techniques and KNN combinations :</span>**","metadata":{}},{"cell_type":"code","source":"pca_df = pd.DataFrame.from_dict(pca_dc_results,orient='index').T\nica_df = pd.DataFrame.from_dict(ica_dc_results,orient='index').T\numap_df = pd.DataFrame.from_dict(umap_dc_results,orient='index').T\nsvd_df = pd.DataFrame.from_dict(svd_dc_results,orient='index').T\nfa_df = pd.DataFrame.from_dict(fa_dc_results,orient='index').T\ntsne_df = pd.DataFrame.from_dict(tsne_dc_results,orient='index').T\nnmf_df = pd.DataFrame.from_dict(nmf_dc_results,orient='index').T\ngrp_df = pd.DataFrame.from_dict(grp_dc_results,orient='index').T\nsrp_df = pd.DataFrame.from_dict(srp_dc_results,orient='index').T\n\ncomparison_df = pd.concat([test_results_df, pca_df, ica_df, umap_df, svd_df,\n                           fa_df, tsne_df, nmf_df, grp_df, srp_df],ignore_index=True)\n\n# Print the resulting DataFrame\ncomparison_df = pd.concat([comparison_df,time_df], axis=1)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"comparison_df","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 12pt\"> Sorting results by best Test Accuracy Score :</span>**","metadata":{}},{"cell_type":"code","source":"comparison_df.sort_values('Test Accuracy',ascending=False)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 12pt\">Sorting by lowest Elapsed time of Dimensionality Reduction Techniques :</span>**","metadata":{}},{"cell_type":"code","source":" comparison_df.sort_values('DC_Time')","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#785a8d;border-radius: 5px;padding: 5px; font-size: 12pt\">Sorting by lowest  Elapsed time Combination of Dimensionality Reduction Techniques and KNN implementation :</span>**","metadata":{}},{"cell_type":"code","source":"comparison_df.sort_values('KNN_DC_Time')","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"<a id=\"7\"></a>\n## <font style=\"font-family: Trebuchet MS; background-color: #371f41; color: white; padding: 12px; line-height: 1;\"> ::: Best Model and Competition Submission ::: </b></font> <a href=\"#top\" class=\"btn btn-primary btn-sm\" role=\"button\" aria-pressed=\"true\" style=\"color:white; background-color: #371f41; margin-left: 30px;\" data-toggle=\"popover\">Go to Table of Contents</a>\n\n___\n\n\n## **<span style=\"color:white;background-color:#371f41;border-radius: 5px;padding: 5px; font-size: 15pt\">Applying Best Combination of Dimensionality Reduction Techniques and KNN implementation</span>**","metadata":{}},{"cell_type":"code","source":"# Applying TruncatedSVD again just to organize the code\nbest_svd = TruncatedSVD(n_components=50, random_state=42)\nX_train_best_svd = best_svd.fit_transform(X_train_scaled)\nX_test_best_svd = best_svd.transform(X_test_scaled)\nX_val_best_svd = best_svd.transform(X_val_scaled)\nX_submission_test_best_svd = best_svd.transform(X_submission_test_scaled)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"best_knn = KNeighborsClassifier(n_neighbors=9 , weights='distance', algorithm='ball_tree',\n                                 leaf_size=34 ,p=2, metric='minkowski')\n\nbest_knn.fit(X_train_best_svd,y_train)\n\n# Evaluate the model on the test set\nbest_y_test_pred = best_knn.predict(X_test_best_svd)\nbest_test_accuracy = accuracy_score(y_test, best_y_test_pred)\n\n# Evaluate the model on the validation set\nbest_y_val_pred = best_knn.predict(X_val_best_svd)\nbest_val_accuracy = accuracy_score(y_val, best_y_val_pred)\n\n# Print the evaluation metrics\nprint(f\"Test Accuracy: {best_test_accuracy}\")\nprint(f\"Validation Accuracy: {best_val_accuracy}\")","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#371f41;border-radius: 5px;padding: 5px; font-size: 15pt\">Competition Submission</span>**","metadata":{}},{"cell_type":"code","source":"predictions = best_knn.predict(X_submission_test_best_svd)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_sample_submission_df['label'] = predictions\ny_sample_submission_df['label'].value_counts().sort_index()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"y_sample_submission_df.to_csv('submission.csv',index=False)\nprint(\"Your submission has been successfully saved\")","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<span style=\"color:white;background-color:#371f41;border-radius: 5px;padding: 5px; font-size: 15pt\">Plotting Submitted Digits Predictions Images</span>**","metadata":{}},{"cell_type":"code","source":"plt.figure(figsize=(20,10))\nfor i in range(40):  \n    plt.subplot(4, 10, i+1)\n    plt.imshow(X_submission_test[i].reshape((28,28)),cmap=plt.cm.binary)\n    plt.title(\"Predicted =%d\" % predictions[i],y=0.9)\n    plt.axis('off')\nplt.subplots_adjust(wspace=0.3, hspace=-0.1)\nplt.show()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## **<center><span style=\"color:white;background-color:#000000;border-radius: 5px;padding: 5px; font-size: 18pt\">If you find this notebook useful or interesting, please, support with an upvote :)</span></center>**\n\n* If you have read this far, I want to thank you for taking the time to engage with my work.\n* Whether you found it insightful or not, or if you noticed areas where I could improve or mistakes i have done, I will be grateful for your feedback.\n* I believe that your feedback is essential for my personal and professional growth, and I would welcome any comments or suggestions that you may have.\n\n__If you enjoyed reading this kernal and are curious about my other work, I invite you to check out my other projects:__\n\n\n**1. [Complete Hyperparameter Tuning 10 Classification Machine Learning Algorithms to Predict Heart Disease ](https://www.kaggle.com/code/amiribrahimtaj/complete-customer-churn-4-tree-based-algorithms/notebook)**\n\n**2. [Time Series - Sales Forecasting - Hyperparameter Tuning 4 evaluated Models ](https://www.kaggle.com/code/amiribrahimtaj/complete-4-time-series-forecasting-tuned-models)**\n\n**3. [Market Segmentation Clustering - K-Means - Mean Shift - Hierarchical - DBSCAN - Spectral - Affinity Propagation](https://www.kaggle.com/code/amiribrahimtaj/market-segmentation-6-tuned-clustering-algorithms)**\n\n**4. [A/B Testing : Which version of the website should you use?](https://www.kaggle.com/code/amiribrahimtaj/a-b-which-version-of-the-website-should-you-use)**\n\n**5. [E-Commerce Sales Exploratory Data Analysis - Time Series Forecasting - Part 2](https://www.kaggle.com/code/amiribrahimtaj/e-commerce-sales-eda-time-series-forecasting-part2)**\n\n**6. [E-Commerce Price Predictive Modeling using 12 Regression Machine Learning Algorithms Full Explained with Complete Hyperparameter Tuning - Part 3](https://www.kaggle.com/code/amiribrahimtaj/complete-tutorial-tuning-12-regression-algorithms)**\n\n**7. [Customer Churn Prediction - Feature Selection and Hyperparameter Tuning - 4 Tree Based Algorithms ](https://www.kaggle.com/code/amiribrahimtaj/complete-customer-churn-4-tree-based-algorithms)**\n\n\n> **Follow me on [LinkedIn](https://www.linkedin.com/in/amiribrahimtag/) and [kaggle](https://www.kaggle.com/amiribrahimtaj) for upcoming tutorials and projects**\n\n__Finally, I will appreciate your upvote and thank you again for your support.__\n","metadata":{}}]}