├── README.md ├── LICENSE └── Logistic-Regression-v1.ipynb /README.md: -------------------------------------------------------------------------------- 1 | # Logistic-Regression 2 | After completing this lab you will be able to: * Use Logistic Regression for classification * Preprocess data for modeling * Implement Logistic regression on real world data 3 | 4 | 5 | ## Logistic Regression for Telco Customer Churn Prediction 6 | 7 | This Jupyter notebook implements a complete **Logistic Regression classification model** to predict customer churn in the telecommunications industry using Python and scikit-learn. The project demonstrates end-to-end machine learning workflow including data preprocessing, model training, evaluation, and feature importance analysis.[1] 8 | 9 | ### Project Overview 10 | 11 | This implementation focuses on predicting which customers are likely to leave a telecommunications company's land-line business for cable competitors. The model achieves approximately **66% accuracy** through logistic regression classification and provides insights into the key factors influencing customer churn.[1] 12 | 13 | ### Dataset 14 | 15 | The project uses the **Telco Customer Churn dataset**, a hypothetical dataset representing telecommunications customer data. The dataset contains 200 customer records with 28 features including:[1] 16 | 17 | **Demographic Features:** 18 | - tenure: Length of customer relationship 19 | - age: Customer age 20 | - address: Years at current address 21 | - income: Annual income 22 | - ed: Education level (1-5 scale) 23 | - employ: Years with current employer 24 | 25 | **Service Features:** 26 | - equip: Equipment rental (binary) 27 | - callcard: Calling card service (binary) 28 | - wireless: Wireless service (binary) 29 | 30 | **Target Variable:** 31 | - churn: Customer churn indicator (0=retained, 1=churned)[1] 32 | 33 | The dataset exhibits a **29% churn rate** (58 out of 200 customers), providing a realistic business scenario for classification modeling.[1] 34 | 35 | ### Key Features 36 | 37 | **Data Preprocessing:** 38 | - Feature selection and subset extraction 39 | - Data type conversion for target variable 40 | - **StandardScaler normalization** for feature scaling 41 | - Train-test split (80-20 ratio) with random state for reproducibility[1] 42 | 43 | **Model Implementation:** 44 | - Logistic Regression classifier from scikit-learn 45 | - Model training on standardized features 46 | - Prediction and probability estimation 47 | - **Log-loss evaluation metric** for model performance assessment[1] 48 | 49 | **Analysis & Visualization:** 50 | - Feature coefficient visualization using horizontal bar plots 51 | - Interpretation of coefficient magnitudes and directions 52 | - Probability-based prediction analysis 53 | - Feature importance ranking for business insights[1] 54 | 55 | ### Technical Stack 56 | 57 | - **Python 3.12.4** 58 | - **pandas 2.2.3**: Data manipulation and analysis 59 | - **numpy 2.2.0**: Numerical computing 60 | - **scikit-learn 1.6.0**: Machine learning algorithms and preprocessing 61 | - **matplotlib 3.9.3**: Data visualization[1] 62 | 63 | ### Model Performance 64 | 65 | The trained model achieves a **log-loss of approximately 0.626** on the test set, indicating reasonable predictive performance. Log-loss (logarithmic loss or binary cross-entropy) measures the performance of a classification model where predictions are probability values between 0 and 1 - lower values indicate better model performance.[1] 66 | 67 | ### Feature Importance Insights 68 | 69 | The model provides interpretable coefficients for each feature, where: 70 | - **Large positive coefficients** indicate that increases in the feature value lead to higher churn probability 71 | - **Large negative coefficients** indicate that increases in the feature value lead to lower churn probability 72 | - **Small absolute values** suggest weaker influence on churn prediction[1] 73 | 74 | The visualization clearly shows which customer attributes have the strongest impact on churn decisions, enabling data-driven retention strategies.[1] 75 | 76 | ### Usage 77 | 78 | The notebook provides a complete, executable workflow that can be run sequentially from top to bottom. Key sections include:[1] 79 | 80 | 1. Library installation and imports 81 | 2. Data loading and exploration 82 | 3. Feature engineering and preprocessing 83 | 4. Data standardization 84 | 5. Train-test splitting 85 | 6. Model training 86 | 7. Prediction generation 87 | 8. Performance evaluation 88 | 9. Feature coefficient analysis and visualization 89 | 90 | ### Practice Exercises 91 | 92 | The notebook includes hands-on exercises to explore model behavior with different feature combinations: 93 | - Adding individual features (callcard, wireless) 94 | - Removing features (equip, income, employ) 95 | - Analyzing the impact on log-loss performance[1] 96 | 97 | ### Learning Objectives 98 | 99 | After completing this implementation, users will be able to: 100 | - Use Logistic Regression for binary classification tasks 101 | - Preprocess data for machine learning modeling 102 | - Implement end-to-end ML pipelines 103 | - Evaluate model performance using appropriate metrics 104 | - Interpret model coefficients for business insights[1] 105 | 106 | ### Estimated Completion Time 107 | 108 | **60 minutes**[1] 109 | 110 | ### Applications 111 | 112 | This implementation is ideal for: 113 | - Customer churn prediction in telecommunications and subscription-based businesses 114 | - Educational purposes for learning classification techniques 115 | - Foundation for more advanced customer retention models 116 | - Business intelligence and customer analytics projects[1] 117 | 118 | ## Dataset link: https://www.kaggle.com/datasets/ezzaldeenesmail/telco-churn-dataset/data 119 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | Apache License 2 | Version 2.0, January 2004 3 | http://www.apache.org/licenses/ 4 | 5 | TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION 6 | 7 | 1. Definitions. 8 | 9 | "License" shall mean the terms and conditions for use, reproduction, 10 | and distribution as defined by Sections 1 through 9 of this document. 11 | 12 | "Licensor" shall mean the copyright owner or entity authorized by 13 | the copyright owner that is granting the License. 14 | 15 | "Legal Entity" shall mean the union of the acting entity and all 16 | other entities that control, are controlled by, or are under common 17 | control with that entity. For the purposes of this definition, 18 | "control" means (i) the power, direct or indirect, to cause the 19 | direction or management of such entity, whether by contract or 20 | otherwise, or (ii) ownership of fifty percent (50%) or more of the 21 | outstanding shares, or (iii) beneficial ownership of such entity. 22 | 23 | "You" (or "Your") shall mean an individual or Legal Entity 24 | exercising permissions granted by this License. 25 | 26 | "Source" form shall mean the preferred form for making modifications, 27 | including but not limited to software source code, documentation 28 | source, and configuration files. 29 | 30 | "Object" form shall mean any form resulting from mechanical 31 | transformation or translation of a Source form, including but 32 | not limited to compiled object code, generated documentation, 33 | and conversions to other media types. 34 | 35 | "Work" shall mean the work of authorship, whether in Source or 36 | Object form, made available under the License, as indicated by a 37 | copyright notice that is included in or attached to the work 38 | (an example is provided in the Appendix below). 39 | 40 | "Derivative Works" shall mean any work, whether in Source or Object 41 | form, that is based on (or derived from) the Work and for which the 42 | editorial revisions, annotations, elaborations, or other modifications 43 | represent, as a whole, an original work of authorship. For the purposes 44 | of this License, Derivative Works shall not include works that remain 45 | separable from, or merely link (or bind by name) to the interfaces of, 46 | the Work and Derivative Works thereof. 47 | 48 | "Contribution" shall mean any work of authorship, including 49 | the original version of the Work and any modifications or additions 50 | to that Work or Derivative Works thereof, that is intentionally 51 | submitted to Licensor for inclusion in the Work by the copyright owner 52 | or by an individual or Legal Entity authorized to submit on behalf of 53 | the copyright owner. For the purposes of this definition, "submitted" 54 | means any form of electronic, verbal, or written communication sent 55 | to the Licensor or its representatives, including but not limited to 56 | communication on electronic mailing lists, source code control systems, 57 | and issue tracking systems that are managed by, or on behalf of, the 58 | Licensor for the purpose of discussing and improving the Work, but 59 | excluding communication that is conspicuously marked or otherwise 60 | designated in writing by the copyright owner as "Not a Contribution." 61 | 62 | "Contributor" shall mean Licensor and any individual or Legal Entity 63 | on behalf of whom a Contribution has been received by Licensor and 64 | subsequently incorporated within the Work. 65 | 66 | 2. Grant of Copyright License. Subject to the terms and conditions of 67 | this License, each Contributor hereby grants to You a perpetual, 68 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 69 | copyright license to reproduce, prepare Derivative Works of, 70 | publicly display, publicly perform, sublicense, and distribute the 71 | Work and such Derivative Works in Source or Object form. 72 | 73 | 3. Grant of Patent License. Subject to the terms and conditions of 74 | this License, each Contributor hereby grants to You a perpetual, 75 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 76 | (except as stated in this section) patent license to make, have made, 77 | use, offer to sell, sell, import, and otherwise transfer the Work, 78 | where such license applies only to those patent claims licensable 79 | by such Contributor that are necessarily infringed by their 80 | Contribution(s) alone or by combination of their Contribution(s) 81 | with the Work to which such Contribution(s) was submitted. If You 82 | institute patent litigation against any entity (including a 83 | cross-claim or counterclaim in a lawsuit) alleging that the Work 84 | or a Contribution incorporated within the Work constitutes direct 85 | or contributory patent infringement, then any patent licenses 86 | granted to You under this License for that Work shall terminate 87 | as of the date such litigation is filed. 88 | 89 | 4. Redistribution. You may reproduce and distribute copies of the 90 | Work or Derivative Works thereof in any medium, with or without 91 | modifications, and in Source or Object form, provided that You 92 | meet the following conditions: 93 | 94 | (a) You must give any other recipients of the Work or 95 | Derivative Works a copy of this License; and 96 | 97 | (b) You must cause any modified files to carry prominent notices 98 | stating that You changed the files; and 99 | 100 | (c) You must retain, in the Source form of any Derivative Works 101 | that You distribute, all copyright, patent, trademark, and 102 | attribution notices from the Source form of the Work, 103 | excluding those notices that do not pertain to any part of 104 | the Derivative Works; and 105 | 106 | (d) If the Work includes a "NOTICE" text file as part of its 107 | distribution, then any Derivative Works that You distribute must 108 | include a readable copy of the attribution notices contained 109 | within such NOTICE file, excluding those notices that do not 110 | pertain to any part of the Derivative Works, in at least one 111 | of the following places: within a NOTICE text file distributed 112 | as part of the Derivative Works; within the Source form or 113 | documentation, if provided along with the Derivative Works; or, 114 | within a display generated by the Derivative Works, if and 115 | wherever such third-party notices normally appear. The contents 116 | of the NOTICE file are for informational purposes only and 117 | do not modify the License. You may add Your own attribution 118 | notices within Derivative Works that You distribute, alongside 119 | or as an addendum to the NOTICE text from the Work, provided 120 | that such additional attribution notices cannot be construed 121 | as modifying the License. 122 | 123 | You may add Your own copyright statement to Your modifications and 124 | may provide additional or different license terms and conditions 125 | for use, reproduction, or distribution of Your modifications, or 126 | for any such Derivative Works as a whole, provided Your use, 127 | reproduction, and distribution of the Work otherwise complies with 128 | the conditions stated in this License. 129 | 130 | 5. Submission of Contributions. Unless You explicitly state otherwise, 131 | any Contribution intentionally submitted for inclusion in the Work 132 | by You to the Licensor shall be under the terms and conditions of 133 | this License, without any additional terms or conditions. 134 | Notwithstanding the above, nothing herein shall supersede or modify 135 | the terms of any separate license agreement you may have executed 136 | with Licensor regarding such Contributions. 137 | 138 | 6. Trademarks. This License does not grant permission to use the trade 139 | names, trademarks, service marks, or product names of the Licensor, 140 | except as required for reasonable and customary use in describing the 141 | origin of the Work and reproducing the content of the NOTICE file. 142 | 143 | 7. Disclaimer of Warranty. Unless required by applicable law or 144 | agreed to in writing, Licensor provides the Work (and each 145 | Contributor provides its Contributions) on an "AS IS" BASIS, 146 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or 147 | implied, including, without limitation, any warranties or conditions 148 | of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A 149 | PARTICULAR PURPOSE. You are solely responsible for determining the 150 | appropriateness of using or redistributing the Work and assume any 151 | risks associated with Your exercise of permissions under this License. 152 | 153 | 8. Limitation of Liability. In no event and under no legal theory, 154 | whether in tort (including negligence), contract, or otherwise, 155 | unless required by applicable law (such as deliberate and grossly 156 | negligent acts) or agreed to in writing, shall any Contributor be 157 | liable to You for damages, including any direct, indirect, special, 158 | incidental, or consequential damages of any character arising as a 159 | result of this License or out of the use or inability to use the 160 | Work (including but not limited to damages for loss of goodwill, 161 | work stoppage, computer failure or malfunction, or any and all 162 | other commercial damages or losses), even if such Contributor 163 | has been advised of the possibility of such damages. 164 | 165 | 9. Accepting Warranty or Additional Liability. While redistributing 166 | the Work or Derivative Works thereof, You may choose to offer, 167 | and charge a fee for, acceptance of support, warranty, indemnity, 168 | or other liability obligations and/or rights consistent with this 169 | License. However, in accepting such obligations, You may act only 170 | on Your own behalf and on Your sole responsibility, not on behalf 171 | of any other Contributor, and only if You agree to indemnify, 172 | defend, and hold each Contributor harmless for any liability 173 | incurred by, or claims asserted against, such Contributor by reason 174 | of your accepting any such warranty or additional liability. 175 | 176 | END OF TERMS AND CONDITIONS 177 | 178 | APPENDIX: How to apply the Apache License to your work. 179 | 180 | To apply the Apache License to your work, attach the following 181 | boilerplate notice, with the fields enclosed by brackets "[]" 182 | replaced with your own identifying information. (Don't include 183 | the brackets!) The text should be enclosed in the appropriate 184 | comment syntax for the file format. We also recommend that a 185 | file or class name and description of purpose be included on the 186 | same "printed page" as the copyright notice for easier 187 | identification within third-party archives. 188 | 189 | Copyright [yyyy] [name of copyright owner] 190 | 191 | Licensed under the Apache License, Version 2.0 (the "License"); 192 | you may not use this file except in compliance with the License. 193 | You may obtain a copy of the License at 194 | 195 | http://www.apache.org/licenses/LICENSE-2.0 196 | 197 | Unless required by applicable law or agreed to in writing, software 198 | distributed under the License is distributed on an "AS IS" BASIS, 199 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 200 | See the License for the specific language governing permissions and 201 | limitations under the License. 202 | -------------------------------------------------------------------------------- /Logistic-Regression-v1.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "markdown", 5 | "id": "fc302825-d4a6-4659-b19d-c4f28a5c3c39", 6 | "metadata": {}, 7 | "source": [ 8 | "\n", 9 | "# Logistic Regression with Python\n", 10 | "\n", 11 | "\n", 12 | "Estimated time needed: **60** minutes\n", 13 | " \n", 14 | "\n", 15 | "## Objectives\n", 16 | "\n", 17 | "After completing this lab you will be able to:\n", 18 | "\n", 19 | "* Use Logistic Regression for classification\n", 20 | "* Preprocess data for modeling\n", 21 | "* Implement Logistic regression on real world data\n" 22 | ] 23 | }, 24 | { 25 | "cell_type": "markdown", 26 | "id": "6b5bca0c-87c7-4445-b2d9-28c2aaf03e08", 27 | "metadata": {}, 28 | "source": [ 29 | "## Install and import the required libraries\n", 30 | "Make sure the required libraries are available by executing the cell below.\n" 31 | ] 32 | }, 33 | { 34 | "cell_type": "code", 35 | "execution_count": null, 36 | "id": "6eadbb65-60c6-47ae-8249-a5a3b12ce36f", 37 | "metadata": {}, 38 | "outputs": [], 39 | "source": [ 40 | "!pip install numpy==2.2.0\n", 41 | "!pip install pandas==2.2.3\n", 42 | "!pip install scikit-learn==1.6.0\n", 43 | "!pip install matplotlib==3.9.3" 44 | ] 45 | }, 46 | { 47 | "cell_type": "markdown", 48 | "id": "28fa82bb-3a10-4e45-8c0d-18e2aa5df684", 49 | "metadata": {}, 50 | "source": [ 51 | "Let's first import required libraries:\n" 52 | ] 53 | }, 54 | { 55 | "cell_type": "markdown", 56 | "metadata": {}, 57 | "source": [ 58 | "import pandas as pd\n", 59 | "import numpy as np\n", 60 | "\n", 61 | "sklearn.model_selection import train_test_split\n", 62 | "\n", 63 | "from sklearn.linear_model import LogisticRegression\n", 64 | "\n", 65 | "from sklearn.preprocessing import StandardScaler\n", 66 | "\n", 67 | "from sklearn.metrics import log_loss\n", 68 | "\n", 69 | "import matplotlib.pyplot as plt\n", 70 | "\n", 71 | "%matplotlib inline \n", 72 | "\n", 73 | "import warnings\n", 74 | "warnings.filterwarnings('ignore')" 75 | ] 76 | }, 77 | { 78 | "cell_type": "code", 79 | "execution_count": null, 80 | "id": "e2c8ace9-9735-40d0-97e6-c97c2445cf76", 81 | "metadata": {}, 82 | "outputs": [], 83 | "source": [ 84 | "import pandas as pd\n", 85 | "import numpy as np\n", 86 | "\n", 87 | "from sklearn.model_selection import train_test_split\n", 88 | "from sklearn.linear_model import LogisticRegression\n", 89 | "from sklearn.preprocessing import StandardScaler\n", 90 | "\n", 91 | "from sklearn.metrics import log_loss\n", 92 | "import matplotlib.pyplot as plt\n", 93 | "\n", 94 | "%matplotlib inline \n", 95 | "\n", 96 | "import warnings\n", 97 | "warnings.filterwarnings('ignore')" 98 | ] 99 | }, 100 | { 101 | "cell_type": "markdown", 102 | "id": "582c85ac-44ea-4ba4-8539-475dee3fbefd", 103 | "metadata": {}, 104 | "source": [ 105 | "## Classification with Logistic Regression\n", 106 | "\n", 107 | "### Scenario\n", 108 | "Assume that you are working for a telecommunications company which is concerned about the number of customers leaving their land-line business for cable competitors. They need to understand who is more likely to leave the company.\n" 109 | ] 110 | }, 111 | { 112 | "cell_type": "markdown", 113 | "id": "2907e2b3-e31b-48c7-b79e-30a9d1f65b29", 114 | "metadata": {}, 115 | "source": [ 116 | "### Load the Telco Churn data \n", 117 | "Telco Churn is a hypothetical data file that concerns a telecommunications company's efforts to reduce turnover in its customer base. Each case corresponds to a separate customer and it records various demographic and service usage information. Before you can work with the data, you must use the URL to get the ChurnData.csv.\n" 118 | ] 119 | }, 120 | { 121 | "cell_type": "markdown", 122 | "id": "77c9cb61-fe06-461d-b05d-16d4b74fdacf", 123 | "metadata": {}, 124 | "source": [ 125 | "### About the dataset\n", 126 | "We will use a telecommunications dataset for predicting customer churn. This is a historical customer dataset where each row represents one customer. The data is relatively easy to understand, and you may uncover insights you can use immediately. Typically it is less expensive to keep customers than acquire new ones, so the focus of this analysis is to predict the customers who will stay with the company. \n", 127 | "

\n", 128 | "This data set provides you information about customer preferences, services opted, personal details, etc. which helps you predict customer churn.\n" 129 | ] 130 | }, 131 | { 132 | "cell_type": "markdown", 133 | "id": "4b0e1cd2-d983-4226-8594-0f92c153c1df", 134 | "metadata": {}, 135 | "source": [ 136 | "### Load Data from URL\n" 137 | ] 138 | }, 139 | { 140 | "cell_type": "code", 141 | "execution_count": null, 142 | "id": "0535963e-0efc-41e9-90b6-6987857af230", 143 | "metadata": {}, 144 | "outputs": [ 145 | { 146 | "data": { 147 | "text/html": [ 148 | "
\n", 149 | "\n", 162 | "\n", 163 | " \n", 164 | " \n", 165 | " \n", 166 | " \n", 167 | " \n", 168 | " \n", 169 | " \n", 170 | " \n", 171 | " \n", 172 | " \n", 173 | " \n", 174 | " \n", 175 | " \n", 176 | " \n", 177 | " \n", 178 | " \n", 179 | " \n", 180 | " \n", 181 | " \n", 182 | " \n", 183 | " \n", 184 | " \n", 185 | " \n", 186 | " \n", 187 | " \n", 188 | " \n", 189 | " \n", 190 | " \n", 191 | " \n", 192 | " \n", 193 | " \n", 194 | " \n", 195 | " \n", 196 | " \n", 197 | " \n", 198 | " \n", 199 | " \n", 200 | " \n", 201 | " \n", 202 | " \n", 203 | " \n", 204 | " \n", 205 | " \n", 206 | " \n", 207 | " \n", 208 | " \n", 209 | " \n", 210 | " \n", 211 | " \n", 212 | " \n", 213 | " \n", 214 | " \n", 215 | " \n", 216 | " \n", 217 | " \n", 218 | " \n", 219 | " \n", 220 | " \n", 221 | " \n", 222 | " \n", 223 | " \n", 224 | " \n", 225 | " \n", 226 | " \n", 227 | " \n", 228 | " \n", 229 | " \n", 230 | " \n", 231 | " \n", 232 | " \n", 233 | " \n", 234 | " \n", 235 | " \n", 236 | " \n", 237 | " \n", 238 | " \n", 239 | " \n", 240 | " \n", 241 | " \n", 242 | " \n", 243 | " \n", 244 | " \n", 245 | " \n", 246 | " \n", 247 | " \n", 248 | " \n", 249 | " \n", 250 | " \n", 251 | " \n", 252 | " \n", 253 | " \n", 254 | " \n", 255 | " \n", 256 | " \n", 257 | " \n", 258 | " \n", 259 | " \n", 260 | " \n", 261 | " \n", 262 | " \n", 263 | " \n", 264 | " \n", 265 | " \n", 266 | " \n", 267 | " \n", 268 | " \n", 269 | " \n", 270 | " \n", 271 | " \n", 272 | " \n", 273 | " \n", 274 | " \n", 275 | " \n", 276 | " \n", 277 | " \n", 278 | " \n", 279 | " \n", 280 | " \n", 281 | " \n", 282 | " \n", 283 | " \n", 284 | " \n", 285 | " \n", 286 | " \n", 287 | " \n", 288 | " \n", 289 | " \n", 290 | " \n", 291 | " \n", 292 | " \n", 293 | " \n", 294 | " \n", 295 | " \n", 296 | " \n", 297 | " \n", 298 | " \n", 299 | " \n", 300 | " \n", 301 | " \n", 302 | " \n", 303 | " \n", 304 | " \n", 305 | " \n", 306 | " \n", 307 | " \n", 308 | " \n", 309 | " \n", 310 | " \n", 311 | " \n", 312 | " \n", 313 | " \n", 314 | " \n", 315 | " \n", 316 | " \n", 317 | " \n", 318 | " \n", 319 | " \n", 320 | " \n", 321 | " \n", 322 | " \n", 323 | " \n", 324 | " \n", 325 | " \n", 326 | " \n", 327 | " \n", 328 | " \n", 329 | " \n", 330 | " \n", 331 | " \n", 332 | " \n", 333 | " \n", 334 | " \n", 335 | " \n", 336 | " \n", 337 | " \n", 338 | " \n", 339 | " \n", 340 | " \n", 341 | " \n", 342 | " \n", 343 | " \n", 344 | " \n", 345 | " \n", 346 | " \n", 347 | " \n", 348 | " \n", 349 | " \n", 350 | " \n", 351 | " \n", 352 | " \n", 353 | " \n", 354 | " \n", 355 | " \n", 356 | " \n", 357 | " \n", 358 | " \n", 359 | " \n", 360 | " \n", 361 | " \n", 362 | " \n", 363 | " \n", 364 | " \n", 365 | " \n", 366 | " \n", 367 | " \n", 368 | " \n", 369 | " \n", 370 | " \n", 371 | " \n", 372 | " \n", 373 | " \n", 374 | " \n", 375 | " \n", 376 | " \n", 377 | " \n", 378 | " \n", 379 | " \n", 380 | " \n", 381 | " \n", 382 | " \n", 383 | " \n", 384 | " \n", 385 | " \n", 386 | " \n", 387 | " \n", 388 | " \n", 389 | " \n", 390 | " \n", 391 | " \n", 392 | " \n", 393 | " \n", 394 | " \n", 395 | " \n", 396 | " \n", 397 | " \n", 398 | " \n", 399 | " \n", 400 | " \n", 401 | " \n", 402 | " \n", 403 | " \n", 404 | " \n", 405 | " \n", 406 | " \n", 407 | " \n", 408 | " \n", 409 | " \n", 410 | " \n", 411 | " \n", 412 | " \n", 413 | " \n", 414 | " \n", 415 | " \n", 416 | " \n", 417 | " \n", 418 | " \n", 419 | " \n", 420 | " \n", 421 | " \n", 422 | " \n", 423 | " \n", 424 | " \n", 425 | " \n", 426 | " \n", 427 | " \n", 428 | " \n", 429 | " \n", 430 | " \n", 431 | " \n", 432 | " \n", 433 | " \n", 434 | " \n", 435 | " \n", 436 | " \n", 437 | " \n", 438 | " \n", 439 | " \n", 440 | " \n", 441 | " \n", 442 | " \n", 443 | " \n", 444 | " \n", 445 | " \n", 446 | " \n", 447 | " \n", 448 | " \n", 449 | " \n", 450 | " \n", 451 | " \n", 452 | " \n", 453 | " \n", 454 | " \n", 455 | "
tenureageaddressincomeedemployequipcallcardwirelesslongmon...pagerinternetcallwaitconferebillloglonglogtolllninccustcatchurn
011.033.07.0136.05.05.00.01.01.04.40...1.00.01.01.00.01.4823.0334.9134.01.0
133.033.012.033.02.00.00.00.00.09.45...0.00.00.00.00.02.2463.2403.4971.01.0
223.030.09.030.01.02.00.00.00.06.30...0.00.00.01.00.01.8413.2403.4013.00.0
338.035.05.076.02.010.01.01.01.06.05...1.01.01.01.01.01.8003.8074.3314.00.0
47.035.014.080.02.015.00.01.00.07.10...0.00.01.01.00.01.9603.0914.3823.00.0
..................................................................
19555.044.024.083.01.023.00.01.00.017.35...0.00.00.01.00.02.8543.1994.4193.00.0
19634.023.03.024.01.07.00.01.00.06.00...0.00.01.01.00.01.7923.3323.1783.00.0
1976.032.010.047.01.010.00.01.00.03.85...0.00.01.01.00.01.3483.1683.8503.00.0
19824.030.00.025.04.05.00.01.01.08.70...1.01.01.01.01.02.1633.8663.2194.01.0
19961.050.016.0190.02.022.01.01.01.016.85...0.01.00.00.01.02.8243.2405.2472.00.0
\n", 456 | "

200 rows × 28 columns

\n", 457 | "
" 458 | ], 459 | "text/plain": [ 460 | " tenure age address income ed employ equip callcard wireless \\\n", 461 | "0 11.0 33.0 7.0 136.0 5.0 5.0 0.0 1.0 1.0 \n", 462 | "1 33.0 33.0 12.0 33.0 2.0 0.0 0.0 0.0 0.0 \n", 463 | "2 23.0 30.0 9.0 30.0 1.0 2.0 0.0 0.0 0.0 \n", 464 | "3 38.0 35.0 5.0 76.0 2.0 10.0 1.0 1.0 1.0 \n", 465 | "4 7.0 35.0 14.0 80.0 2.0 15.0 0.0 1.0 0.0 \n", 466 | ".. ... ... ... ... ... ... ... ... ... \n", 467 | "195 55.0 44.0 24.0 83.0 1.0 23.0 0.0 1.0 0.0 \n", 468 | "196 34.0 23.0 3.0 24.0 1.0 7.0 0.0 1.0 0.0 \n", 469 | "197 6.0 32.0 10.0 47.0 1.0 10.0 0.0 1.0 0.0 \n", 470 | "198 24.0 30.0 0.0 25.0 4.0 5.0 0.0 1.0 1.0 \n", 471 | "199 61.0 50.0 16.0 190.0 2.0 22.0 1.0 1.0 1.0 \n", 472 | "\n", 473 | " longmon ... pager internet callwait confer ebill loglong logtoll \\\n", 474 | "0 4.40 ... 1.0 0.0 1.0 1.0 0.0 1.482 3.033 \n", 475 | "1 9.45 ... 0.0 0.0 0.0 0.0 0.0 2.246 3.240 \n", 476 | "2 6.30 ... 0.0 0.0 0.0 1.0 0.0 1.841 3.240 \n", 477 | "3 6.05 ... 1.0 1.0 1.0 1.0 1.0 1.800 3.807 \n", 478 | "4 7.10 ... 0.0 0.0 1.0 1.0 0.0 1.960 3.091 \n", 479 | ".. ... ... ... ... ... ... ... ... ... \n", 480 | "195 17.35 ... 0.0 0.0 0.0 1.0 0.0 2.854 3.199 \n", 481 | "196 6.00 ... 0.0 0.0 1.0 1.0 0.0 1.792 3.332 \n", 482 | "197 3.85 ... 0.0 0.0 1.0 1.0 0.0 1.348 3.168 \n", 483 | "198 8.70 ... 1.0 1.0 1.0 1.0 1.0 2.163 3.866 \n", 484 | "199 16.85 ... 0.0 1.0 0.0 0.0 1.0 2.824 3.240 \n", 485 | "\n", 486 | " lninc custcat churn \n", 487 | "0 4.913 4.0 1.0 \n", 488 | "1 3.497 1.0 1.0 \n", 489 | "2 3.401 3.0 0.0 \n", 490 | "3 4.331 4.0 0.0 \n", 491 | "4 4.382 3.0 0.0 \n", 492 | ".. ... ... ... \n", 493 | "195 4.419 3.0 0.0 \n", 494 | "196 3.178 3.0 0.0 \n", 495 | "197 3.850 3.0 0.0 \n", 496 | "198 3.219 4.0 1.0 \n", 497 | "199 5.247 2.0 0.0 \n", 498 | "\n", 499 | "[200 rows x 28 columns]" 500 | ] 501 | }, 502 | "execution_count": 2, 503 | "metadata": {}, 504 | "output_type": "execute_result" 505 | } 506 | ], 507 | "source": [ 508 | "# churn_df = pd.read_csv(\"ChurnData.csv\")\n", 509 | "# url = \"https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBMDeveloperSkillsNetwork-ML0101EN-SkillsNetwork/labs/Module%203/data/ChurnData.csv\"\n", 510 | "url = \"Telco_Churn_dataset.csv\"\n", 511 | "churn_df = pd.read_csv(url)\n", 512 | "churn_df" 513 | ] 514 | }, 515 | { 516 | "cell_type": "markdown", 517 | "id": "4401b830-4ab6-4e91-bf1f-bd697d713408", 518 | "metadata": {}, 519 | "source": [ 520 | "Let's select some features for the modeling. Also, we change the target data type to be an integer, as it is a requirement by the scikit-learn algorithm:\n" 521 | ] 522 | }, 523 | { 524 | "cell_type": "markdown", 525 | "id": "437fb566-3724-4dcc-8367-9894fd8241f9", 526 | "metadata": {}, 527 | "source": [ 528 | "## Data Preprocessing\n" 529 | ] 530 | }, 531 | { 532 | "cell_type": "markdown", 533 | "id": "853fe610-ca22-4e8c-a00c-bce324cb6164", 534 | "metadata": {}, 535 | "source": [ 536 | "For this lab, we can use a subset of the fields available to develop out model. Let us assume that the fields we use are 'tenure', 'age', 'address', 'income', 'ed', 'employ', 'equip' and of course 'churn'.\n" 537 | ] 538 | }, 539 | { 540 | "cell_type": "code", 541 | "execution_count": 3, 542 | "metadata": {}, 543 | "outputs": [ 544 | { 545 | "name": "stdout", 546 | "output_type": "stream", 547 | "text": [ 548 | "\n", 549 | "RangeIndex: 200 entries, 0 to 199\n", 550 | "Data columns (total 8 columns):\n", 551 | " # Column Non-Null Count Dtype \n", 552 | "--- ------ -------------- ----- \n", 553 | " 0 tenure 200 non-null float64\n", 554 | " 1 age 200 non-null float64\n", 555 | " 2 address 200 non-null float64\n", 556 | " 3 income 200 non-null float64\n", 557 | " 4 ed 200 non-null float64\n", 558 | " 5 employ 200 non-null float64\n", 559 | " 6 equip 200 non-null float64\n", 560 | " 7 churn 200 non-null float64\n", 561 | "dtypes: float64(8)\n", 562 | "memory usage: 12.6 KB\n" 563 | ] 564 | } 565 | ], 566 | "source": [ 567 | "churn_df = churn_df[['tenure', 'age', 'address', 'income', 'ed', 'employ', 'equip', 'churn']]\n", 568 | "churn_df.info()" 569 | ] 570 | }, 571 | { 572 | "cell_type": "code", 573 | "execution_count": 4, 574 | "metadata": {}, 575 | "outputs": [ 576 | { 577 | "data": { 578 | "text/html": [ 579 | "
\n", 580 | "\n", 593 | "\n", 594 | " \n", 595 | " \n", 596 | " \n", 597 | " \n", 598 | " \n", 599 | " \n", 600 | " \n", 601 | " \n", 602 | " \n", 603 | " \n", 604 | " \n", 605 | " \n", 606 | " \n", 607 | " \n", 608 | " \n", 609 | " \n", 610 | " \n", 611 | " \n", 612 | " \n", 613 | " \n", 614 | " \n", 615 | " \n", 616 | " \n", 617 | " \n", 618 | " \n", 619 | " \n", 620 | " \n", 621 | " \n", 622 | " \n", 623 | " \n", 624 | " \n", 625 | " \n", 626 | " \n", 627 | " \n", 628 | " \n", 629 | " \n", 630 | " \n", 631 | " \n", 632 | " \n", 633 | " \n", 634 | " \n", 635 | " \n", 636 | " \n", 637 | " \n", 638 | " \n", 639 | " \n", 640 | " \n", 641 | " \n", 642 | " \n", 643 | " \n", 644 | " \n", 645 | " \n", 646 | " \n", 647 | " \n", 648 | " \n", 649 | " \n", 650 | " \n", 651 | " \n", 652 | " \n", 653 | " \n", 654 | " \n", 655 | " \n", 656 | " \n", 657 | " \n", 658 | " \n", 659 | " \n", 660 | " \n", 661 | " \n", 662 | " \n", 663 | " \n", 664 | "
tenureageaddressincomeedemployequipchurn
011.033.07.0136.05.05.00.01.0
133.033.012.033.02.00.00.01.0
223.030.09.030.01.02.00.00.0
338.035.05.076.02.010.01.00.0
47.035.014.080.02.015.00.00.0
\n", 665 | "
" 666 | ], 667 | "text/plain": [ 668 | " tenure age address income ed employ equip churn\n", 669 | "0 11.0 33.0 7.0 136.0 5.0 5.0 0.0 1.0\n", 670 | "1 33.0 33.0 12.0 33.0 2.0 0.0 0.0 1.0\n", 671 | "2 23.0 30.0 9.0 30.0 1.0 2.0 0.0 0.0\n", 672 | "3 38.0 35.0 5.0 76.0 2.0 10.0 1.0 0.0\n", 673 | "4 7.0 35.0 14.0 80.0 2.0 15.0 0.0 0.0" 674 | ] 675 | }, 676 | "execution_count": 4, 677 | "metadata": {}, 678 | "output_type": "execute_result" 679 | } 680 | ], 681 | "source": [ 682 | "churn_df.head()" 683 | ] 684 | }, 685 | { 686 | "cell_type": "code", 687 | "execution_count": 5, 688 | "id": "732b526c-6548-489a-8529-6102d0622a1d", 689 | "metadata": {}, 690 | "outputs": [ 691 | { 692 | "data": { 693 | "text/html": [ 694 | "
\n", 695 | "\n", 708 | "\n", 709 | " \n", 710 | " \n", 711 | " \n", 712 | " \n", 713 | " \n", 714 | " \n", 715 | " \n", 716 | " \n", 717 | " \n", 718 | " \n", 719 | " \n", 720 | " \n", 721 | " \n", 722 | " \n", 723 | " \n", 724 | " \n", 725 | " \n", 726 | " \n", 727 | " \n", 728 | " \n", 729 | " \n", 730 | " \n", 731 | " \n", 732 | " \n", 733 | " \n", 734 | " \n", 735 | " \n", 736 | " \n", 737 | " \n", 738 | " \n", 739 | " \n", 740 | " \n", 741 | " \n", 742 | " \n", 743 | " \n", 744 | " \n", 745 | " \n", 746 | " \n", 747 | " \n", 748 | " \n", 749 | " \n", 750 | " \n", 751 | " \n", 752 | " \n", 753 | " \n", 754 | " \n", 755 | " \n", 756 | " \n", 757 | " \n", 758 | " \n", 759 | " \n", 760 | " \n", 761 | " \n", 762 | " \n", 763 | " \n", 764 | " \n", 765 | " \n", 766 | " \n", 767 | " \n", 768 | " \n", 769 | " \n", 770 | " \n", 771 | " \n", 772 | " \n", 773 | " \n", 774 | " \n", 775 | " \n", 776 | " \n", 777 | " \n", 778 | " \n", 779 | " \n", 780 | " \n", 781 | " \n", 782 | " \n", 783 | " \n", 784 | " \n", 785 | " \n", 786 | " \n", 787 | " \n", 788 | " \n", 789 | " \n", 790 | " \n", 791 | " \n", 792 | " \n", 793 | " \n", 794 | " \n", 795 | " \n", 796 | " \n", 797 | " \n", 798 | " \n", 799 | " \n", 800 | " \n", 801 | " \n", 802 | " \n", 803 | " \n", 804 | " \n", 805 | " \n", 806 | " \n", 807 | " \n", 808 | " \n", 809 | " \n", 810 | " \n", 811 | " \n", 812 | " \n", 813 | " \n", 814 | " \n", 815 | " \n", 816 | " \n", 817 | " \n", 818 | " \n", 819 | " \n", 820 | " \n", 821 | " \n", 822 | " \n", 823 | " \n", 824 | " \n", 825 | " \n", 826 | " \n", 827 | " \n", 828 | " \n", 829 | " \n", 830 | " \n", 831 | " \n", 832 | " \n", 833 | " \n", 834 | " \n", 835 | " \n", 836 | " \n", 837 | " \n", 838 | " \n", 839 | " \n", 840 | " \n", 841 | " \n", 842 | " \n", 843 | " \n", 844 | " \n", 845 | "
tenureageaddressincomeedemployequipchurn
011.033.07.0136.05.05.00.01
133.033.012.033.02.00.00.01
223.030.09.030.01.02.00.00
338.035.05.076.02.010.01.00
47.035.014.080.02.015.00.00
...........................
19555.044.024.083.01.023.00.00
19634.023.03.024.01.07.00.00
1976.032.010.047.01.010.00.00
19824.030.00.025.04.05.00.01
19961.050.016.0190.02.022.01.00
\n", 846 | "

200 rows × 8 columns

\n", 847 | "
" 848 | ], 849 | "text/plain": [ 850 | " tenure age address income ed employ equip churn\n", 851 | "0 11.0 33.0 7.0 136.0 5.0 5.0 0.0 1\n", 852 | "1 33.0 33.0 12.0 33.0 2.0 0.0 0.0 1\n", 853 | "2 23.0 30.0 9.0 30.0 1.0 2.0 0.0 0\n", 854 | "3 38.0 35.0 5.0 76.0 2.0 10.0 1.0 0\n", 855 | "4 7.0 35.0 14.0 80.0 2.0 15.0 0.0 0\n", 856 | ".. ... ... ... ... ... ... ... ...\n", 857 | "195 55.0 44.0 24.0 83.0 1.0 23.0 0.0 0\n", 858 | "196 34.0 23.0 3.0 24.0 1.0 7.0 0.0 0\n", 859 | "197 6.0 32.0 10.0 47.0 1.0 10.0 0.0 0\n", 860 | "198 24.0 30.0 0.0 25.0 4.0 5.0 0.0 1\n", 861 | "199 61.0 50.0 16.0 190.0 2.0 22.0 1.0 0\n", 862 | "\n", 863 | "[200 rows x 8 columns]" 864 | ] 865 | }, 866 | "execution_count": 5, 867 | "metadata": {}, 868 | "output_type": "execute_result" 869 | } 870 | ], 871 | "source": [ 872 | "churn_df['churn'] = churn_df['churn'].astype('int')\n", 873 | "churn_df" 874 | ] 875 | }, 876 | { 877 | "cell_type": "markdown", 878 | "id": "5ca4c39a-e8e7-4ede-8c1c-54360507a566", 879 | "metadata": {}, 880 | "source": [ 881 | "For modeling the input fields X and the target field y need to be fixed. Since that the target to be predicted is 'churn', the data under this field will be stored under the variable 'y'. We may use any combination or all of the remaining fields as the input. Store these values in the variable 'X'.\n" 882 | ] 883 | }, 884 | { 885 | "cell_type": "code", 886 | "execution_count": null, 887 | "metadata": {}, 888 | "outputs": [ 889 | { 890 | "data": { 891 | "text/html": [ 892 | "
\n", 893 | "\n", 906 | "\n", 907 | " \n", 908 | " \n", 909 | " \n", 910 | " \n", 911 | " \n", 912 | " \n", 913 | " \n", 914 | " \n", 915 | " \n", 916 | " \n", 917 | " \n", 918 | " \n", 919 | " \n", 920 | " \n", 921 | " \n", 922 | " \n", 923 | " \n", 924 | " \n", 925 | " \n", 926 | " \n", 927 | " \n", 928 | " \n", 929 | " \n", 930 | " \n", 931 | " \n", 932 | " \n", 933 | " \n", 934 | " \n", 935 | " \n", 936 | " \n", 937 | " \n", 938 | " \n", 939 | " \n", 940 | " \n", 941 | " \n", 942 | " \n", 943 | " \n", 944 | " \n", 945 | " \n", 946 | " \n", 947 | " \n", 948 | " \n", 949 | " \n", 950 | " \n", 951 | " \n", 952 | " \n", 953 | " \n", 954 | " \n", 955 | " \n", 956 | " \n", 957 | " \n", 958 | " \n", 959 | " \n", 960 | " \n", 961 | " \n", 962 | " \n", 963 | " \n", 964 | " \n", 965 | " \n", 966 | " \n", 967 | " \n", 968 | " \n", 969 | " \n", 970 | " \n", 971 | " \n", 972 | " \n", 973 | " \n", 974 | " \n", 975 | " \n", 976 | " \n", 977 | " \n", 978 | " \n", 979 | " \n", 980 | " \n", 981 | " \n", 982 | " \n", 983 | " \n", 984 | " \n", 985 | " \n", 986 | " \n", 987 | " \n", 988 | " \n", 989 | " \n", 990 | " \n", 991 | " \n", 992 | " \n", 993 | " \n", 994 | " \n", 995 | " \n", 996 | " \n", 997 | " \n", 998 | " \n", 999 | " \n", 1000 | " \n", 1001 | " \n", 1002 | " \n", 1003 | " \n", 1004 | " \n", 1005 | " \n", 1006 | " \n", 1007 | " \n", 1008 | " \n", 1009 | " \n", 1010 | " \n", 1011 | " \n", 1012 | " \n", 1013 | " \n", 1014 | " \n", 1015 | " \n", 1016 | " \n", 1017 | " \n", 1018 | " \n", 1019 | " \n", 1020 | " \n", 1021 | " \n", 1022 | " \n", 1023 | " \n", 1024 | " \n", 1025 | " \n", 1026 | " \n", 1027 | " \n", 1028 | " \n", 1029 | " \n", 1030 | " \n", 1031 | "
tenureageaddressincomeedemployequip
011.033.07.0136.05.05.00.0
133.033.012.033.02.00.00.0
223.030.09.030.01.02.00.0
338.035.05.076.02.010.01.0
47.035.014.080.02.015.00.0
........................
19555.044.024.083.01.023.00.0
19634.023.03.024.01.07.00.0
1976.032.010.047.01.010.00.0
19824.030.00.025.04.05.00.0
19961.050.016.0190.02.022.01.0
\n", 1032 | "

200 rows × 7 columns

\n", 1033 | "
" 1034 | ], 1035 | "text/plain": [ 1036 | " tenure age address income ed employ equip\n", 1037 | "0 11.0 33.0 7.0 136.0 5.0 5.0 0.0\n", 1038 | "1 33.0 33.0 12.0 33.0 2.0 0.0 0.0\n", 1039 | "2 23.0 30.0 9.0 30.0 1.0 2.0 0.0\n", 1040 | "3 38.0 35.0 5.0 76.0 2.0 10.0 1.0\n", 1041 | "4 7.0 35.0 14.0 80.0 2.0 15.0 0.0\n", 1042 | ".. ... ... ... ... ... ... ...\n", 1043 | "195 55.0 44.0 24.0 83.0 1.0 23.0 0.0\n", 1044 | "196 34.0 23.0 3.0 24.0 1.0 7.0 0.0\n", 1045 | "197 6.0 32.0 10.0 47.0 1.0 10.0 0.0\n", 1046 | "198 24.0 30.0 0.0 25.0 4.0 5.0 0.0\n", 1047 | "199 61.0 50.0 16.0 190.0 2.0 22.0 1.0\n", 1048 | "\n", 1049 | "[200 rows x 7 columns]" 1050 | ] 1051 | }, 1052 | "execution_count": 13, 1053 | "metadata": {}, 1054 | "output_type": "execute_result" 1055 | } 1056 | ], 1057 | "source": [ 1058 | "XX = churn_df.drop('churn', axis=1)\n", 1059 | "yy = churn_df['churn']\n", 1060 | "XX" 1061 | ] 1062 | }, 1063 | { 1064 | "cell_type": "code", 1065 | "execution_count": 6, 1066 | "id": "9a4afcab-17e1-450e-b6df-e14d5d94eb26", 1067 | "metadata": {}, 1068 | "outputs": [ 1069 | { 1070 | "data": { 1071 | "text/plain": [ 1072 | "array([[ 11., 33., 7., 136., 5., 5., 0.],\n", 1073 | " [ 33., 33., 12., 33., 2., 0., 0.],\n", 1074 | " [ 23., 30., 9., 30., 1., 2., 0.],\n", 1075 | " [ 38., 35., 5., 76., 2., 10., 1.],\n", 1076 | " [ 7., 35., 14., 80., 2., 15., 0.]])" 1077 | ] 1078 | }, 1079 | "execution_count": 6, 1080 | "metadata": {}, 1081 | "output_type": "execute_result" 1082 | } 1083 | ], 1084 | "source": [ 1085 | "# X = XX.to_numpy()\n", 1086 | "X = np.asarray(churn_df[['tenure', 'age', 'address', 'income', 'ed', 'employ', 'equip']])\n", 1087 | "X[0:5] #print the first 5 values" 1088 | ] 1089 | }, 1090 | { 1091 | "cell_type": "code", 1092 | "execution_count": 7, 1093 | "id": "b5a6b91c-f3ee-499f-a844-7eabae0d9408", 1094 | "metadata": {}, 1095 | "outputs": [ 1096 | { 1097 | "data": { 1098 | "text/plain": [ 1099 | "array([1, 1, 0, 0, 0])" 1100 | ] 1101 | }, 1102 | "execution_count": 7, 1103 | "metadata": {}, 1104 | "output_type": "execute_result" 1105 | } 1106 | ], 1107 | "source": [ 1108 | "y = np.asarray(churn_df['churn'])\n", 1109 | "y[0:5] #print the first 5 values" 1110 | ] 1111 | }, 1112 | { 1113 | "cell_type": "markdown", 1114 | "id": "026bedf6-e320-4e58-8655-3a4ba111139f", 1115 | "metadata": {}, 1116 | "source": [ 1117 | "It is also a norm to standardize or normalize the dataset in order to have all the features at the same scale. This helps the model learn faster and improves the model performance. We may make use of StandardScalar function in the Scikit-Learn library.\n" 1118 | ] 1119 | }, 1120 | { 1121 | "cell_type": "code", 1122 | "execution_count": 15, 1123 | "id": "71cb424f-c701-42b8-8426-db16a8f3d1cd", 1124 | "metadata": {}, 1125 | "outputs": [ 1126 | { 1127 | "data": { 1128 | "text/plain": [ 1129 | "array([[-1.13518441, -0.62595491, -0.4588971 , 0.4751423 , 1.6961288 ,\n", 1130 | " -0.58477841, -0.85972695],\n", 1131 | " [-0.11604313, -0.62595491, 0.03454064, -0.32886061, -0.6433592 ,\n", 1132 | " -1.14437497, -0.85972695],\n", 1133 | " [-0.57928917, -0.85594447, -0.261522 , -0.35227817, -1.42318853,\n", 1134 | " -0.92053635, -0.85972695],\n", 1135 | " [ 0.11557989, -0.47262854, -0.65627219, 0.00679109, -0.6433592 ,\n", 1136 | " -0.02518185, 1.16316 ],\n", 1137 | " [-1.32048283, -0.47262854, 0.23191574, 0.03801451, -0.6433592 ,\n", 1138 | " 0.53441472, -0.85972695]])" 1139 | ] 1140 | }, 1141 | "execution_count": 15, 1142 | "metadata": {}, 1143 | "output_type": "execute_result" 1144 | } 1145 | ], 1146 | "source": [ 1147 | "X_norm = StandardScaler().fit(X).transform(X)\n", 1148 | "X_norm[0:5]" 1149 | ] 1150 | }, 1151 | { 1152 | "cell_type": "markdown", 1153 | "id": "180bb824-9b40-4a53-808f-e10bec6d677e", 1154 | "metadata": {}, 1155 | "source": [ 1156 | "### Splitting the dataset\n" 1157 | ] 1158 | }, 1159 | { 1160 | "cell_type": "markdown", 1161 | "id": "8971e904-5c1c-4694-bef0-a7b62fd94b1f", 1162 | "metadata": {}, 1163 | "source": [ 1164 | "The trained model has to be tested and evaluated on data which has not been used during training. Therefore, it is required to separate a part of the data for testing and the remaining for training. For this, we may make use of the train_test_split function in the scikit-learn library.\n" 1165 | ] 1166 | }, 1167 | { 1168 | "cell_type": "code", 1169 | "execution_count": 16, 1170 | "id": "cdbe3030-85a2-4656-abc0-1f1f7ba041e5", 1171 | "metadata": {}, 1172 | "outputs": [], 1173 | "source": [ 1174 | "X_train, X_test, y_train, y_test = train_test_split( X_norm, y, test_size=0.2, random_state=4)" 1175 | ] 1176 | }, 1177 | { 1178 | "cell_type": "markdown", 1179 | "id": "e873eb89-30ba-4b75-b5db-2d6c653d5ff9", 1180 | "metadata": {}, 1181 | "source": [ 1182 | "## Logistic Regression Classifier modeling\n" 1183 | ] 1184 | }, 1185 | { 1186 | "cell_type": "markdown", 1187 | "id": "f925b359-dffb-42ed-ae01-23cf4c08b031", 1188 | "metadata": {}, 1189 | "source": [ 1190 | "Let's build the model using __LogisticRegression__ from the Scikit-learn package and fit our model with train data set.\n" 1191 | ] 1192 | }, 1193 | { 1194 | "cell_type": "code", 1195 | "execution_count": 17, 1196 | "id": "b57308e1-5e95-4e07-a021-ebf72842e21c", 1197 | "metadata": {}, 1198 | "outputs": [], 1199 | "source": [ 1200 | "LR = LogisticRegression().fit(X_train,y_train)" 1201 | ] 1202 | }, 1203 | { 1204 | "cell_type": "markdown", 1205 | "id": "89a92441-51e5-4dc9-ae90-fa22e5d7ff1f", 1206 | "metadata": {}, 1207 | "source": [ 1208 | "Fitting, or in simple terms training, gives us a model that has now learnt from the traning data and can be used to predict the output variable. Let us predict the churn parameter for the test data set.\n" 1209 | ] 1210 | }, 1211 | { 1212 | "cell_type": "code", 1213 | "execution_count": 19, 1214 | "id": "e2253e7b-5519-4be6-97b6-34663295740d", 1215 | "metadata": {}, 1216 | "outputs": [ 1217 | { 1218 | "data": { 1219 | "text/plain": [ 1220 | "(array([0, 0, 0, 0, 0, 0, 0, 0, 1, 0]), array([1, 1, 0, 0, 0, 0, 0, 0, 0, 0]))" 1221 | ] 1222 | }, 1223 | "execution_count": 19, 1224 | "metadata": {}, 1225 | "output_type": "execute_result" 1226 | } 1227 | ], 1228 | "source": [ 1229 | "yhat = LR.predict(X_test)\n", 1230 | "yhat[:10], y[:10]" 1231 | ] 1232 | }, 1233 | { 1234 | "cell_type": "markdown", 1235 | "id": "df6d5069-6b71-4f0a-b120-d6bd097f8d80", 1236 | "metadata": {}, 1237 | "source": [ 1238 | "To understand this prediction, we can also have a look at the prediction probability of data point of the test data set. Use the function __predict_proba__ , we can get the probability of each class. The first column is the probability of the record belonging to class 0, and second column that of class 1. Note that the class prediction system uses the threshold for class prediction as 0.5. This means that the class predicted is the one which is most likely.\n" 1239 | ] 1240 | }, 1241 | { 1242 | "cell_type": "code", 1243 | "execution_count": 20, 1244 | "id": "dedd0a86-c822-4d86-a5a9-765340e2437b", 1245 | "metadata": {}, 1246 | "outputs": [ 1247 | { 1248 | "data": { 1249 | "text/plain": [ 1250 | "array([[0.74643946, 0.25356054],\n", 1251 | " [0.92667894, 0.07332106],\n", 1252 | " [0.83442627, 0.16557373],\n", 1253 | " [0.94600618, 0.05399382],\n", 1254 | " [0.84325532, 0.15674468],\n", 1255 | " [0.71448367, 0.28551633],\n", 1256 | " [0.77076426, 0.22923574],\n", 1257 | " [0.90955642, 0.09044358],\n", 1258 | " [0.26152115, 0.73847885],\n", 1259 | " [0.94900731, 0.05099269]])" 1260 | ] 1261 | }, 1262 | "execution_count": 20, 1263 | "metadata": {}, 1264 | "output_type": "execute_result" 1265 | } 1266 | ], 1267 | "source": [ 1268 | "yhat_prob = LR.predict_proba(X_test)\n", 1269 | "yhat_prob[:10]" 1270 | ] 1271 | }, 1272 | { 1273 | "cell_type": "markdown", 1274 | "id": "ee9c420b-79c0-4195-ba05-dc5505716296", 1275 | "metadata": {}, 1276 | "source": [ 1277 | "Since the purpose here is to predict the 1 class more acccurately, you can also examine what role each input feature has to play in the prediction of the 1 class. Consider the code below.\n" 1278 | ] 1279 | }, 1280 | { 1281 | "cell_type": "code", 1282 | "execution_count": null, 1283 | "id": "b9cc2884-c83b-4fa7-8d84-8ad6fa15ed22", 1284 | "metadata": {}, 1285 | "outputs": [ 1286 | { 1287 | "data": { 1288 | "image/png": "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", 1289 | "text/plain": [ 1290 | "
" 1291 | ] 1292 | }, 1293 | "metadata": {}, 1294 | "output_type": "display_data" 1295 | } 1296 | ], 1297 | "source": [ 1298 | "coefficients = pd.Series(LR.coef_[0], index=churn_df.columns[:-1])\n", 1299 | "coefficients.sort_values().plot(kind='barh')\n", 1300 | "plt.title(\"Feature Coefficients in Logistic Regression Churn Model\")\n", 1301 | "plt.xlabel(\"Coefficient Value\")\n", 1302 | "plt.show()" 1303 | ] 1304 | }, 1305 | { 1306 | "cell_type": "markdown", 1307 | "id": "96a198b8-c505-40c9-8c6f-a24e3d15b860", 1308 | "metadata": {}, 1309 | "source": [ 1310 | "Large positive value of LR Coefficient for a given field indicates that increase in this parameter will lead to better chance of a positive, i.e. 1 class. A large negative value indicates the opposite, which means that an increase in this parameter will lead to poorer chance of a positive class. A lower absolute value indicates weaker affect of the change in that field on the predicted class. Let us examine this with the following exercises. \n" 1311 | ] 1312 | }, 1313 | { 1314 | "cell_type": "markdown", 1315 | "id": "74ec8246-0f37-41ff-8995-5691d9c8ebd2", 1316 | "metadata": {}, 1317 | "source": [ 1318 | "## Performance Evaluation\n" 1319 | ] 1320 | }, 1321 | { 1322 | "cell_type": "markdown", 1323 | "id": "74d44824-d0fc-4c8a-a45f-132fadef6488", 1324 | "metadata": {}, 1325 | "source": [ 1326 | "Once the predictions have been generated, it becomes prudent to evaluate the performance of the model in predicting the target variable. Let us evaluate the log-loss value.\n", 1327 | "\n", 1328 | "### log loss\n", 1329 | "\n", 1330 | "Log loss (Logarithmic loss), also known as Binary Cross entropy loss, is a function that generates a loss value based on the class wise prediction probabilities and the actual class labels. The lower the log loss value, the better the model is considered to be.\n" 1331 | ] 1332 | }, 1333 | { 1334 | "cell_type": "code", 1335 | "execution_count": 28, 1336 | "id": "8dba58f1-502d-4e32-93fc-b8d7903eb36e", 1337 | "metadata": {}, 1338 | "outputs": [ 1339 | { 1340 | "data": { 1341 | "text/plain": [ 1342 | "0.6257718410257236" 1343 | ] 1344 | }, 1345 | "execution_count": 28, 1346 | "metadata": {}, 1347 | "output_type": "execute_result" 1348 | } 1349 | ], 1350 | "source": [ 1351 | "log_loss(y_test, yhat_prob)" 1352 | ] 1353 | }, 1354 | { 1355 | "cell_type": "code", 1356 | "execution_count": 32, 1357 | "metadata": {}, 1358 | "outputs": [ 1359 | { 1360 | "data": { 1361 | "text/html": [ 1362 | "
\n", 1363 | "\n", 1376 | "\n", 1377 | " \n", 1378 | " \n", 1379 | " \n", 1380 | " \n", 1381 | " \n", 1382 | " \n", 1383 | " \n", 1384 | " \n", 1385 | " \n", 1386 | " \n", 1387 | " \n", 1388 | " \n", 1389 | " \n", 1390 | " \n", 1391 | " \n", 1392 | " \n", 1393 | " \n", 1394 | " \n", 1395 | " \n", 1396 | " \n", 1397 | " \n", 1398 | " \n", 1399 | " \n", 1400 | " \n", 1401 | " \n", 1402 | " \n", 1403 | " \n", 1404 | " \n", 1405 | " \n", 1406 | " \n", 1407 | " \n", 1408 | " \n", 1409 | " \n", 1410 | " \n", 1411 | " \n", 1412 | " \n", 1413 | " \n", 1414 | " \n", 1415 | " \n", 1416 | " \n", 1417 | " \n", 1418 | " \n", 1419 | " \n", 1420 | " \n", 1421 | " \n", 1422 | " \n", 1423 | " \n", 1424 | " \n", 1425 | " \n", 1426 | " \n", 1427 | " \n", 1428 | " \n", 1429 | " \n", 1430 | " \n", 1431 | " \n", 1432 | " \n", 1433 | " \n", 1434 | " \n", 1435 | " \n", 1436 | " \n", 1437 | " \n", 1438 | " \n", 1439 | " \n", 1440 | " \n", 1441 | " \n", 1442 | " \n", 1443 | " \n", 1444 | " \n", 1445 | " \n", 1446 | " \n", 1447 | " \n", 1448 | " \n", 1449 | " \n", 1450 | " \n", 1451 | " \n", 1452 | " \n", 1453 | " \n", 1454 | " \n", 1455 | " \n", 1456 | " \n", 1457 | " \n", 1458 | " \n", 1459 | " \n", 1460 | " \n", 1461 | " \n", 1462 | " \n", 1463 | " \n", 1464 | " \n", 1465 | " \n", 1466 | " \n", 1467 | " \n", 1468 | " \n", 1469 | " \n", 1470 | " \n", 1471 | " \n", 1472 | " \n", 1473 | " \n", 1474 | " \n", 1475 | " \n", 1476 | " \n", 1477 | " \n", 1478 | " \n", 1479 | " \n", 1480 | " \n", 1481 | " \n", 1482 | " \n", 1483 | " \n", 1484 | " \n", 1485 | " \n", 1486 | " \n", 1487 | " \n", 1488 | " \n", 1489 | " \n", 1490 | " \n", 1491 | " \n", 1492 | " \n", 1493 | " \n", 1494 | " \n", 1495 | " \n", 1496 | " \n", 1497 | " \n", 1498 | " \n", 1499 | " \n", 1500 | " \n", 1501 | " \n", 1502 | " \n", 1503 | " \n", 1504 | " \n", 1505 | " \n", 1506 | " \n", 1507 | " \n", 1508 | " \n", 1509 | " \n", 1510 | " \n", 1511 | " \n", 1512 | " \n", 1513 | " \n", 1514 | " \n", 1515 | " \n", 1516 | " \n", 1517 | " \n", 1518 | " \n", 1519 | " \n", 1520 | " \n", 1521 | " \n", 1522 | " \n", 1523 | " \n", 1524 | " \n", 1525 | "
tenureageaddressincomeedemployequipcallcardchurn
011.033.07.0136.05.05.00.01.01
133.033.012.033.02.00.00.00.01
223.030.09.030.01.02.00.00.00
338.035.05.076.02.010.01.01.00
47.035.014.080.02.015.00.01.00
..............................
19555.044.024.083.01.023.00.01.00
19634.023.03.024.01.07.00.01.00
1976.032.010.047.01.010.00.01.00
19824.030.00.025.04.05.00.01.01
19961.050.016.0190.02.022.01.01.00
\n", 1526 | "

200 rows × 9 columns

\n", 1527 | "
" 1528 | ], 1529 | "text/plain": [ 1530 | " tenure age address income ed employ equip callcard churn\n", 1531 | "0 11.0 33.0 7.0 136.0 5.0 5.0 0.0 1.0 1\n", 1532 | "1 33.0 33.0 12.0 33.0 2.0 0.0 0.0 0.0 1\n", 1533 | "2 23.0 30.0 9.0 30.0 1.0 2.0 0.0 0.0 0\n", 1534 | "3 38.0 35.0 5.0 76.0 2.0 10.0 1.0 1.0 0\n", 1535 | "4 7.0 35.0 14.0 80.0 2.0 15.0 0.0 1.0 0\n", 1536 | ".. ... ... ... ... ... ... ... ... ...\n", 1537 | "195 55.0 44.0 24.0 83.0 1.0 23.0 0.0 1.0 0\n", 1538 | "196 34.0 23.0 3.0 24.0 1.0 7.0 0.0 1.0 0\n", 1539 | "197 6.0 32.0 10.0 47.0 1.0 10.0 0.0 1.0 0\n", 1540 | "198 24.0 30.0 0.0 25.0 4.0 5.0 0.0 1.0 1\n", 1541 | "199 61.0 50.0 16.0 190.0 2.0 22.0 1.0 1.0 0\n", 1542 | "\n", 1543 | "[200 rows x 9 columns]" 1544 | ] 1545 | }, 1546 | "execution_count": 32, 1547 | "metadata": {}, 1548 | "output_type": "execute_result" 1549 | } 1550 | ], 1551 | "source": [ 1552 | "df = churn_df[['tenure', 'age', 'address', 'income', 'ed', 'employ', 'equip', 'callcard', 'churn']]\n", 1553 | "df['churn'] = df['churn'].astype(int)\n", 1554 | "df" 1555 | ] 1556 | }, 1557 | { 1558 | "cell_type": "code", 1559 | "execution_count": 33, 1560 | "metadata": {}, 1561 | "outputs": [ 1562 | { 1563 | "data": { 1564 | "text/plain": [ 1565 | "array([[ 11., 33., 7., 136., 5., 5., 0., 1.],\n", 1566 | " [ 33., 33., 12., 33., 2., 0., 0., 0.],\n", 1567 | " [ 23., 30., 9., 30., 1., 2., 0., 0.],\n", 1568 | " [ 38., 35., 5., 76., 2., 10., 1., 1.],\n", 1569 | " [ 7., 35., 14., 80., 2., 15., 0., 1.],\n", 1570 | " [ 68., 52., 17., 120., 1., 24., 0., 1.],\n", 1571 | " [ 42., 40., 7., 37., 2., 8., 1., 1.],\n", 1572 | " [ 9., 21., 1., 17., 2., 2., 0., 0.],\n", 1573 | " [ 35., 50., 26., 140., 2., 21., 0., 1.],\n", 1574 | " [ 49., 51., 27., 63., 4., 19., 0., 1.]])" 1575 | ] 1576 | }, 1577 | "execution_count": 33, 1578 | "metadata": {}, 1579 | "output_type": "execute_result" 1580 | } 1581 | ], 1582 | "source": [ 1583 | "X = np.asarray(df.drop('churn', axis=1))\n", 1584 | "X[:10]" 1585 | ] 1586 | }, 1587 | { 1588 | "cell_type": "code", 1589 | "execution_count": 34, 1590 | "metadata": {}, 1591 | "outputs": [ 1592 | { 1593 | "data": { 1594 | "text/plain": [ 1595 | "array([1, 1, 0, 0, 0, 0, 0, 0, 0, 0])" 1596 | ] 1597 | }, 1598 | "execution_count": 34, 1599 | "metadata": {}, 1600 | "output_type": "execute_result" 1601 | } 1602 | ], 1603 | "source": [ 1604 | "y = np.asarray(df['churn'])\n", 1605 | "y[:10]" 1606 | ] 1607 | }, 1608 | { 1609 | "cell_type": "code", 1610 | "execution_count": 35, 1611 | "metadata": {}, 1612 | "outputs": [ 1613 | { 1614 | "data": { 1615 | "text/plain": [ 1616 | "array([[-1.13518441, -0.62595491, -0.4588971 , 0.4751423 , 1.6961288 ,\n", 1617 | " -0.58477841, -0.85972695, 0.64686916],\n", 1618 | " [-0.11604313, -0.62595491, 0.03454064, -0.32886061, -0.6433592 ,\n", 1619 | " -1.14437497, -0.85972695, -1.54590766],\n", 1620 | " [-0.57928917, -0.85594447, -0.261522 , -0.35227817, -1.42318853,\n", 1621 | " -0.92053635, -0.85972695, -1.54590766],\n", 1622 | " [ 0.11557989, -0.47262854, -0.65627219, 0.00679109, -0.6433592 ,\n", 1623 | " -0.02518185, 1.16316 , 0.64686916],\n", 1624 | " [-1.32048283, -0.47262854, 0.23191574, 0.03801451, -0.6433592 ,\n", 1625 | " 0.53441472, -0.85972695, 0.64686916],\n", 1626 | " [ 1.505318 , 0.83064562, 0.52797838, 0.35024865, -1.42318853,\n", 1627 | " 1.54168853, -0.85972695, 0.64686916],\n", 1628 | " [ 0.3008783 , -0.08931261, -0.4588971 , -0.29763719, -0.6433592 ,\n", 1629 | " -0.24902047, 1.16316 , 0.64686916],\n", 1630 | " [-1.22783362, -1.54591314, -1.05102238, -0.45375426, -0.6433592 ,\n", 1631 | " -0.92053635, -0.85972695, -1.54590766],\n", 1632 | " [-0.02339392, 0.67731925, 1.41616631, 0.50636572, -0.6433592 ,\n", 1633 | " 1.2059306 , -0.85972695, 0.64686916],\n", 1634 | " [ 0.62515053, 0.75398243, 1.51485386, -0.094685 , 0.91629947,\n", 1635 | " 0.98209197, -0.85972695, 0.64686916]])" 1636 | ] 1637 | }, 1638 | "execution_count": 35, 1639 | "metadata": {}, 1640 | "output_type": "execute_result" 1641 | } 1642 | ], 1643 | "source": [ 1644 | "X_scal = StandardScaler().fit(X).transform(X)\n", 1645 | "X_scal[:10]" 1646 | ] 1647 | }, 1648 | { 1649 | "cell_type": "code", 1650 | "execution_count": 36, 1651 | "metadata": {}, 1652 | "outputs": [], 1653 | "source": [ 1654 | "X_train, X_test, y_train, y_test = train_test_split(X_scal, y, test_size=0.2, random_state=42)" 1655 | ] 1656 | }, 1657 | { 1658 | "cell_type": "code", 1659 | "execution_count": 37, 1660 | "metadata": {}, 1661 | "outputs": [], 1662 | "source": [ 1663 | "model = LogisticRegression().fit(X_scal, y)" 1664 | ] 1665 | }, 1666 | { 1667 | "cell_type": "code", 1668 | "execution_count": 39, 1669 | "metadata": {}, 1670 | "outputs": [ 1671 | { 1672 | "data": { 1673 | "text/plain": [ 1674 | "(array([1, 1, 0, 0, 0, 1, 0, 1, 0, 0]), array([1, 1, 0, 0, 1, 1, 0, 0, 1, 0]))" 1675 | ] 1676 | }, 1677 | "execution_count": 39, 1678 | "metadata": {}, 1679 | "output_type": "execute_result" 1680 | } 1681 | ], 1682 | "source": [ 1683 | "y_pred = model.predict(X_test)\n", 1684 | "y_pred[:10], y_test[:10]" 1685 | ] 1686 | }, 1687 | { 1688 | "cell_type": "code", 1689 | "execution_count": 40, 1690 | "metadata": {}, 1691 | "outputs": [ 1692 | { 1693 | "data": { 1694 | "text/plain": [ 1695 | "array([[0.40548452, 0.59451548],\n", 1696 | " [0.1892232 , 0.8107768 ],\n", 1697 | " [0.83833063, 0.16166937],\n", 1698 | " [0.80353379, 0.19646621],\n", 1699 | " [0.72392994, 0.27607006],\n", 1700 | " [0.23808895, 0.76191105],\n", 1701 | " [0.98609233, 0.01390767],\n", 1702 | " [0.43214602, 0.56785398],\n", 1703 | " [0.5192971 , 0.4807029 ],\n", 1704 | " [0.95840862, 0.04159138]])" 1705 | ] 1706 | }, 1707 | "execution_count": 40, 1708 | "metadata": {}, 1709 | "output_type": "execute_result" 1710 | } 1711 | ], 1712 | "source": [ 1713 | "y_prop = model.predict_proba(X_test)\n", 1714 | "y_prop[:10]" 1715 | ] 1716 | }, 1717 | { 1718 | "cell_type": "code", 1719 | "execution_count": null, 1720 | "metadata": {}, 1721 | "outputs": [ 1722 | { 1723 | "data": { 1724 | "image/png": "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", 1725 | "text/plain": [ 1726 | "
" 1727 | ] 1728 | }, 1729 | "metadata": {}, 1730 | "output_type": "display_data" 1731 | } 1732 | ], 1733 | "source": [ 1734 | "coefficients = pd.Series(model.coef_[0], index= df.columns[:-1])\n", 1735 | "coefficients.sort_values().plot(kind='barh')\n", 1736 | "plt.title(\"Feature Coefficients in Logistic Regression Churn Model\")\n", 1737 | "plt.ylabel('coefficients values')\n", 1738 | "plt.show()\n" 1739 | ] 1740 | }, 1741 | { 1742 | "cell_type": "code", 1743 | "execution_count": 50, 1744 | "metadata": {}, 1745 | "outputs": [ 1746 | { 1747 | "data": { 1748 | "text/plain": [ 1749 | "0.33478619890420447" 1750 | ] 1751 | }, 1752 | "execution_count": 50, 1753 | "metadata": {}, 1754 | "output_type": "execute_result" 1755 | } 1756 | ], 1757 | "source": [ 1758 | "# log_loss()\n", 1759 | "log_loss(y_test, y_prop)" 1760 | ] 1761 | }, 1762 | { 1763 | "cell_type": "markdown", 1764 | "id": "ba67cb2f-185f-436e-897e-cbf6bfba8d32", 1765 | "metadata": {}, 1766 | "source": [ 1767 | "## Practice Exercises\n", 1768 | "Try to attempt the following questions yourself based on what you learnt in this lab.\n", 1769 | "\n", 1770 | "
\n", 1771 | "\n", 1772 | "a. Let us assume we add the feature 'callcard' to the original set of input features. What will the value of log loss be in this case?\n", 1773 | "
\n", 1774 | "
Hint\n", 1775 | "Reuse all the code statements above after modifying the value of churn_df. Make sure to edit the list of features feeding the variable X. The expected answer is 0.6039104035600186.\n", 1776 | "
\n", 1777 | "\n", 1778 | "b. Let us assume we add the feature 'wireless' to the original set of input features. What will the value of log loss be in this case?\n", 1779 | "
\n", 1780 | "
Hint\n", 1781 | "Reuse all the code statements above after modifying the value of churn_df. Make sure to edit the list of features feeding the variable X. The expected answer is 0.7227054293985518.\n", 1782 | "
\n", 1783 | "\n", 1784 | "c. What happens to the log loss value if we add both \"callcard\" and \"wireless\" to the input features?\n", 1785 | "
\n", 1786 | "
Hint\n", 1787 | "Reuse all the code statements above after modifying the value of churn_df. Make sure to edit the list of features feeding the variable X. The expected answer is 0.7760557225417114\n", 1788 | "
\n", 1789 | "\n", 1790 | "d. What happens to the log loss if we remove the feature 'equip' from the original set of input features?\n", 1791 | "
\n", 1792 | "
Hint\n", 1793 | "Reuse all the code statements above after modifying the value of churn_df Make sure to edit the list of features feeding the variable X. The expected answer is 0.5302427350245369\n", 1794 | "
\n", 1795 | "\n", 1796 | "e. What happens to the log loss if we remove the features 'income' and 'employ' from the original set of input features?\n", 1797 | "
\n", 1798 | "
Hint\n", 1799 | "Reuse all the code statements above after modifying the value of churn_df. Make sure to edit the list of features feeding the variable X. The expected answer is 0.7821990869010905.\n", 1800 | "
\n" 1801 | ] 1802 | }, 1803 | { 1804 | "cell_type": "markdown", 1805 | "id": "6618b207-dcd7-44bc-bd1f-4dbdff27ec70", 1806 | "metadata": {}, 1807 | "source": [ 1808 | "### Congratulations! You're ready to move on to your next lesson.\n", 1809 | " \n", 1810 | " \n", 1811 | "## Author\n", 1812 | "Ezzaldeen Esmail \n", 1813 | "\n", 1814 | "\n" 1815 | ] 1816 | } 1817 | ], 1818 | "metadata": { 1819 | "kernelspec": { 1820 | "display_name": "base", 1821 | "language": "python", 1822 | "name": "python3" 1823 | }, 1824 | "language_info": { 1825 | "codemirror_mode": { 1826 | "name": "ipython", 1827 | "version": 3 1828 | }, 1829 | "file_extension": ".py", 1830 | "mimetype": "text/x-python", 1831 | "name": "python", 1832 | "nbconvert_exporter": "python", 1833 | "pygments_lexer": "ipython3", 1834 | "version": "3.12.4" 1835 | }, 1836 | "prev_pub_hash": "f5074de89ffd23dee06449d9ee52e7c9bcc4951457475d26a88adac5df2bee9d" 1837 | }, 1838 | "nbformat": 4, 1839 | "nbformat_minor": 4 1840 | } 1841 | --------------------------------------------------------------------------------