├── LICENSE ├── Model Interpretation.ipynb ├── Model_Performance.ipynb ├── Predictive analytics.ipynb ├── README.md ├── basetable_ex2_4.csv ├── classification ├── respiration │ ├── DT.ipynb │ ├── knn.ipynb │ └── svm.ipynb └── spill_incidents_analytics │ ├── README.md │ ├── emerson_spill_incidents_eda.ipynb │ ├── emerson_spill_incidents_modelling.ipynb │ └── requirements.txt └── data ├── AUC.JPG ├── PIG_continuous.JPG ├── basetable.csv ├── basetable_ex_2_13.csv ├── basetable_interactions.csv ├── cum_gain.JPG ├── cumulative_gains.JPG ├── cutoff.JPG ├── gifts.csv ├── lift_curve.JPG ├── lift_curve_interpret.JPG ├── living_places.csv └── predictor_insights_graph.JPG /LICENSE: -------------------------------------------------------------------------------- 1 | GNU GENERAL PUBLIC LICENSE 2 | Version 3, 29 June 2007 3 | 4 | Copyright (C) 2007 Free Software Foundation, Inc. 5 | Everyone is permitted to copy and distribute verbatim copies 6 | of this license document, but changing it is not allowed. 7 | 8 | Preamble 9 | 10 | The GNU General Public License is a free, copyleft license for 11 | software and other kinds of works. 12 | 13 | The licenses for most software and other practical works are designed 14 | to take away your freedom to share and change the works. By contrast, 15 | the GNU General Public License is intended to guarantee your freedom to 16 | share and change all versions of a program--to make sure it remains free 17 | software for all its users. We, the Free Software Foundation, use the 18 | GNU General Public License for most of our software; it applies also to 19 | any other work released this way by its authors. You can apply it to 20 | your programs, too. 21 | 22 | When we speak of free software, we are referring to freedom, not 23 | price. Our General Public Licenses are designed to make sure that you 24 | have the freedom to distribute copies of free software (and charge for 25 | them if you wish), that you receive source code or can get it if you 26 | want it, that you can change the software or use pieces of it in new 27 | free programs, and that you know you can do these things. 28 | 29 | To protect your rights, we need to prevent others from denying you 30 | these rights or asking you to surrender the rights. Therefore, you have 31 | certain responsibilities if you distribute copies of the software, or if 32 | you modify it: responsibilities to respect the freedom of others. 33 | 34 | For example, if you distribute copies of such a program, whether 35 | gratis or for a fee, you must pass on to the recipients the same 36 | freedoms that you received. You must make sure that they, too, receive 37 | or can get the source code. And you must show them these terms so they 38 | know their rights. 39 | 40 | Developers that use the GNU GPL protect your rights with two steps: 41 | (1) assert copyright on the software, and (2) offer you this License 42 | giving you legal permission to copy, distribute and/or modify it. 43 | 44 | For the developers' and authors' protection, the GPL clearly explains 45 | that there is no warranty for this free software. For both users' and 46 | authors' sake, the GPL requires that modified versions be marked as 47 | changed, so that their problems will not be attributed erroneously to 48 | authors of previous versions. 49 | 50 | Some devices are designed to deny users access to install or run 51 | modified versions of the software inside them, although the manufacturer 52 | can do so. This is fundamentally incompatible with the aim of 53 | protecting users' freedom to change the software. The systematic 54 | pattern of such abuse occurs in the area of products for individuals to 55 | use, which is precisely where it is most unacceptable. Therefore, we 56 | have designed this version of the GPL to prohibit the practice for those 57 | products. If such problems arise substantially in other domains, we 58 | stand ready to extend this provision to those domains in future versions 59 | of the GPL, as needed to protect the freedom of users. 60 | 61 | Finally, every program is threatened constantly by software patents. 62 | States should not allow patents to restrict development and use of 63 | software on general-purpose computers, but in those that do, we wish to 64 | avoid the special danger that patents applied to a free program could 65 | make it effectively proprietary. To prevent this, the GPL assures that 66 | patents cannot be used to render the program non-free. 67 | 68 | The precise terms and conditions for copying, distribution and 69 | modification follow. 70 | 71 | TERMS AND CONDITIONS 72 | 73 | 0. Definitions. 74 | 75 | "This License" refers to version 3 of the GNU General Public License. 76 | 77 | "Copyright" also means copyright-like laws that apply to other kinds of 78 | works, such as semiconductor masks. 79 | 80 | "The Program" refers to any copyrightable work licensed under this 81 | License. Each licensee is addressed as "you". "Licensees" and 82 | "recipients" may be individuals or organizations. 83 | 84 | To "modify" a work means to copy from or adapt all or part of the work 85 | in a fashion requiring copyright permission, other than the making of an 86 | exact copy. The resulting work is called a "modified version" of the 87 | earlier work or a work "based on" the earlier work. 88 | 89 | A "covered work" means either the unmodified Program or a work based 90 | on the Program. 91 | 92 | To "propagate" a work means to do anything with it that, without 93 | permission, would make you directly or secondarily liable for 94 | infringement under applicable copyright law, except executing it on a 95 | computer or modifying a private copy. Propagation includes copying, 96 | distribution (with or without modification), making available to the 97 | public, and in some countries other activities as well. 98 | 99 | To "convey" a work means any kind of propagation that enables other 100 | parties to make or receive copies. Mere interaction with a user through 101 | a computer network, with no transfer of a copy, is not conveying. 102 | 103 | An interactive user interface displays "Appropriate Legal Notices" 104 | to the extent that it includes a convenient and prominently visible 105 | feature that (1) displays an appropriate copyright notice, and (2) 106 | tells the user that there is no warranty for the work (except to the 107 | extent that warranties are provided), that licensees may convey the 108 | work under this License, and how to view a copy of this License. If 109 | the interface presents a list of user commands or options, such as a 110 | menu, a prominent item in the list meets this criterion. 111 | 112 | 1. Source Code. 113 | 114 | The "source code" for a work means the preferred form of the work 115 | for making modifications to it. "Object code" means any non-source 116 | form of a work. 117 | 118 | A "Standard Interface" means an interface that either is an official 119 | standard defined by a recognized standards body, or, in the case of 120 | interfaces specified for a particular programming language, one that 121 | is widely used among developers working in that language. 122 | 123 | The "System Libraries" of an executable work include anything, other 124 | than the work as a whole, that (a) is included in the normal form of 125 | packaging a Major Component, but which is not part of that Major 126 | Component, and (b) serves only to enable use of the work with that 127 | Major Component, or to implement a Standard Interface for which an 128 | implementation is available to the public in source code form. A 129 | "Major Component", in this context, means a major essential component 130 | (kernel, window system, and so on) of the specific operating system 131 | (if any) on which the executable work runs, or a compiler used to 132 | produce the work, or an object code interpreter used to run it. 133 | 134 | The "Corresponding Source" for a work in object code form means all 135 | the source code needed to generate, install, and (for an executable 136 | work) run the object code and to modify the work, including scripts to 137 | control those activities. However, it does not include the work's 138 | System Libraries, or general-purpose tools or generally available free 139 | programs which are used unmodified in performing those activities but 140 | which are not part of the work. For example, Corresponding Source 141 | includes interface definition files associated with source files for 142 | the work, and the source code for shared libraries and dynamically 143 | linked subprograms that the work is specifically designed to require, 144 | such as by intimate data communication or control flow between those 145 | subprograms and other parts of the work. 146 | 147 | The Corresponding Source need not include anything that users 148 | can regenerate automatically from other parts of the Corresponding 149 | Source. 150 | 151 | The Corresponding Source for a work in source code form is that 152 | same work. 153 | 154 | 2. Basic Permissions. 155 | 156 | All rights granted under this License are granted for the term of 157 | copyright on the Program, and are irrevocable provided the stated 158 | conditions are met. This License explicitly affirms your unlimited 159 | permission to run the unmodified Program. The output from running a 160 | covered work is covered by this License only if the output, given its 161 | content, constitutes a covered work. This License acknowledges your 162 | rights of fair use or other equivalent, as provided by copyright law. 163 | 164 | You may make, run and propagate covered works that you do not 165 | convey, without conditions so long as your license otherwise remains 166 | in force. You may convey covered works to others for the sole purpose 167 | of having them make modifications exclusively for you, or provide you 168 | with facilities for running those works, provided that you comply with 169 | the terms of this License in conveying all material for which you do 170 | not control copyright. Those thus making or running the covered works 171 | for you must do so exclusively on your behalf, under your direction 172 | and control, on terms that prohibit them from making any copies of 173 | your copyrighted material outside their relationship with you. 174 | 175 | Conveying under any other circumstances is permitted solely under 176 | the conditions stated below. Sublicensing is not allowed; section 10 177 | makes it unnecessary. 178 | 179 | 3. Protecting Users' Legal Rights From Anti-Circumvention Law. 180 | 181 | No covered work shall be deemed part of an effective technological 182 | measure under any applicable law fulfilling obligations under article 183 | 11 of the WIPO copyright treaty adopted on 20 December 1996, or 184 | similar laws prohibiting or restricting circumvention of such 185 | measures. 186 | 187 | When you convey a covered work, you waive any legal power to forbid 188 | circumvention of technological measures to the extent such circumvention 189 | is effected by exercising rights under this License with respect to 190 | the covered work, and you disclaim any intention to limit operation or 191 | modification of the work as a means of enforcing, against the work's 192 | users, your or third parties' legal rights to forbid circumvention of 193 | technological measures. 194 | 195 | 4. Conveying Verbatim Copies. 196 | 197 | You may convey verbatim copies of the Program's source code as you 198 | receive it, in any medium, provided that you conspicuously and 199 | appropriately publish on each copy an appropriate copyright notice; 200 | keep intact all notices stating that this License and any 201 | non-permissive terms added in accord with section 7 apply to the code; 202 | keep intact all notices of the absence of any warranty; and give all 203 | recipients a copy of this License along with the Program. 204 | 205 | You may charge any price or no price for each copy that you convey, 206 | and you may offer support or warranty protection for a fee. 207 | 208 | 5. Conveying Modified Source Versions. 209 | 210 | You may convey a work based on the Program, or the modifications to 211 | produce it from the Program, in the form of source code under the 212 | terms of section 4, provided that you also meet all of these conditions: 213 | 214 | a) The work must carry prominent notices stating that you modified 215 | it, and giving a relevant date. 216 | 217 | b) The work must carry prominent notices stating that it is 218 | released under this License and any conditions added under section 219 | 7. This requirement modifies the requirement in section 4 to 220 | "keep intact all notices". 221 | 222 | c) You must license the entire work, as a whole, under this 223 | License to anyone who comes into possession of a copy. This 224 | License will therefore apply, along with any applicable section 7 225 | additional terms, to the whole of the work, and all its parts, 226 | regardless of how they are packaged. This License gives no 227 | permission to license the work in any other way, but it does not 228 | invalidate such permission if you have separately received it. 229 | 230 | d) If the work has interactive user interfaces, each must display 231 | Appropriate Legal Notices; however, if the Program has interactive 232 | interfaces that do not display Appropriate Legal Notices, your 233 | work need not make them do so. 234 | 235 | A compilation of a covered work with other separate and independent 236 | works, which are not by their nature extensions of the covered work, 237 | and which are not combined with it such as to form a larger program, 238 | in or on a volume of a storage or distribution medium, is called an 239 | "aggregate" if the compilation and its resulting copyright are not 240 | used to limit the access or legal rights of the compilation's users 241 | beyond what the individual works permit. Inclusion of a covered work 242 | in an aggregate does not cause this License to apply to the other 243 | parts of the aggregate. 244 | 245 | 6. Conveying Non-Source Forms. 246 | 247 | You may convey a covered work in object code form under the terms 248 | of sections 4 and 5, provided that you also convey the 249 | machine-readable Corresponding Source under the terms of this License, 250 | in one of these ways: 251 | 252 | a) Convey the object code in, or embodied in, a physical product 253 | (including a physical distribution medium), accompanied by the 254 | Corresponding Source fixed on a durable physical medium 255 | customarily used for software interchange. 256 | 257 | b) Convey the object code in, or embodied in, a physical product 258 | (including a physical distribution medium), accompanied by a 259 | written offer, valid for at least three years and valid for as 260 | long as you offer spare parts or customer support for that product 261 | model, to give anyone who possesses the object code either (1) a 262 | copy of the Corresponding Source for all the software in the 263 | product that is covered by this License, on a durable physical 264 | medium customarily used for software interchange, for a price no 265 | more than your reasonable cost of physically performing this 266 | conveying of source, or (2) access to copy the 267 | Corresponding Source from a network server at no charge. 268 | 269 | c) Convey individual copies of the object code with a copy of the 270 | written offer to provide the Corresponding Source. This 271 | alternative is allowed only occasionally and noncommercially, and 272 | only if you received the object code with such an offer, in accord 273 | with subsection 6b. 274 | 275 | d) Convey the object code by offering access from a designated 276 | place (gratis or for a charge), and offer equivalent access to the 277 | Corresponding Source in the same way through the same place at no 278 | further charge. You need not require recipients to copy the 279 | Corresponding Source along with the object code. If the place to 280 | copy the object code is a network server, the Corresponding Source 281 | may be on a different server (operated by you or a third party) 282 | that supports equivalent copying facilities, provided you maintain 283 | clear directions next to the object code saying where to find the 284 | Corresponding Source. Regardless of what server hosts the 285 | Corresponding Source, you remain obligated to ensure that it is 286 | available for as long as needed to satisfy these requirements. 287 | 288 | e) Convey the object code using peer-to-peer transmission, provided 289 | you inform other peers where the object code and Corresponding 290 | Source of the work are being offered to the general public at no 291 | charge under subsection 6d. 292 | 293 | A separable portion of the object code, whose source code is excluded 294 | from the Corresponding Source as a System Library, need not be 295 | included in conveying the object code work. 296 | 297 | A "User Product" is either (1) a "consumer product", which means any 298 | tangible personal property which is normally used for personal, family, 299 | or household purposes, or (2) anything designed or sold for incorporation 300 | into a dwelling. In determining whether a product is a consumer product, 301 | doubtful cases shall be resolved in favor of coverage. For a particular 302 | product received by a particular user, "normally used" refers to a 303 | typical or common use of that class of product, regardless of the status 304 | of the particular user or of the way in which the particular user 305 | actually uses, or expects or is expected to use, the product. A product 306 | is a consumer product regardless of whether the product has substantial 307 | commercial, industrial or non-consumer uses, unless such uses represent 308 | the only significant mode of use of the product. 309 | 310 | "Installation Information" for a User Product means any methods, 311 | procedures, authorization keys, or other information required to install 312 | and execute modified versions of a covered work in that User Product from 313 | a modified version of its Corresponding Source. The information must 314 | suffice to ensure that the continued functioning of the modified object 315 | code is in no case prevented or interfered with solely because 316 | modification has been made. 317 | 318 | If you convey an object code work under this section in, or with, or 319 | specifically for use in, a User Product, and the conveying occurs as 320 | part of a transaction in which the right of possession and use of the 321 | User Product is transferred to the recipient in perpetuity or for a 322 | fixed term (regardless of how the transaction is characterized), the 323 | Corresponding Source conveyed under this section must be accompanied 324 | by the Installation Information. But this requirement does not apply 325 | if neither you nor any third party retains the ability to install 326 | modified object code on the User Product (for example, the work has 327 | been installed in ROM). 328 | 329 | The requirement to provide Installation Information does not include a 330 | requirement to continue to provide support service, warranty, or updates 331 | for a work that has been modified or installed by the recipient, or for 332 | the User Product in which it has been modified or installed. Access to a 333 | network may be denied when the modification itself materially and 334 | adversely affects the operation of the network or violates the rules and 335 | protocols for communication across the network. 336 | 337 | Corresponding Source conveyed, and Installation Information provided, 338 | in accord with this section must be in a format that is publicly 339 | documented (and with an implementation available to the public in 340 | source code form), and must require no special password or key for 341 | unpacking, reading or copying. 342 | 343 | 7. Additional Terms. 344 | 345 | "Additional permissions" are terms that supplement the terms of this 346 | License by making exceptions from one or more of its conditions. 347 | Additional permissions that are applicable to the entire Program shall 348 | be treated as though they were included in this License, to the extent 349 | that they are valid under applicable law. If additional permissions 350 | apply only to part of the Program, that part may be used separately 351 | under those permissions, but the entire Program remains governed by 352 | this License without regard to the additional permissions. 353 | 354 | When you convey a copy of a covered work, you may at your option 355 | remove any additional permissions from that copy, or from any part of 356 | it. (Additional permissions may be written to require their own 357 | removal in certain cases when you modify the work.) You may place 358 | additional permissions on material, added by you to a covered work, 359 | for which you have or can give appropriate copyright permission. 360 | 361 | Notwithstanding any other provision of this License, for material you 362 | add to a covered work, you may (if authorized by the copyright holders of 363 | that material) supplement the terms of this License with terms: 364 | 365 | a) Disclaiming warranty or limiting liability differently from the 366 | terms of sections 15 and 16 of this License; or 367 | 368 | b) Requiring preservation of specified reasonable legal notices or 369 | author attributions in that material or in the Appropriate Legal 370 | Notices displayed by works containing it; or 371 | 372 | c) Prohibiting misrepresentation of the origin of that material, or 373 | requiring that modified versions of such material be marked in 374 | reasonable ways as different from the original version; or 375 | 376 | d) Limiting the use for publicity purposes of names of licensors or 377 | authors of the material; or 378 | 379 | e) Declining to grant rights under trademark law for use of some 380 | trade names, trademarks, or service marks; or 381 | 382 | f) Requiring indemnification of licensors and authors of that 383 | material by anyone who conveys the material (or modified versions of 384 | it) with contractual assumptions of liability to the recipient, for 385 | any liability that these contractual assumptions directly impose on 386 | those licensors and authors. 387 | 388 | All other non-permissive additional terms are considered "further 389 | restrictions" within the meaning of section 10. If the Program as you 390 | received it, or any part of it, contains a notice stating that it is 391 | governed by this License along with a term that is a further 392 | restriction, you may remove that term. If a license document contains 393 | a further restriction but permits relicensing or conveying under this 394 | License, you may add to a covered work material governed by the terms 395 | of that license document, provided that the further restriction does 396 | not survive such relicensing or conveying. 397 | 398 | If you add terms to a covered work in accord with this section, you 399 | must place, in the relevant source files, a statement of the 400 | additional terms that apply to those files, or a notice indicating 401 | where to find the applicable terms. 402 | 403 | Additional terms, permissive or non-permissive, may be stated in the 404 | form of a separately written license, or stated as exceptions; 405 | the above requirements apply either way. 406 | 407 | 8. Termination. 408 | 409 | You may not propagate or modify a covered work except as expressly 410 | provided under this License. Any attempt otherwise to propagate or 411 | modify it is void, and will automatically terminate your rights under 412 | this License (including any patent licenses granted under the third 413 | paragraph of section 11). 414 | 415 | However, if you cease all violation of this License, then your 416 | license from a particular copyright holder is reinstated (a) 417 | provisionally, unless and until the copyright holder explicitly and 418 | finally terminates your license, and (b) permanently, if the copyright 419 | holder fails to notify you of the violation by some reasonable means 420 | prior to 60 days after the cessation. 421 | 422 | Moreover, your license from a particular copyright holder is 423 | reinstated permanently if the copyright holder notifies you of the 424 | violation by some reasonable means, this is the first time you have 425 | received notice of violation of this License (for any work) from that 426 | copyright holder, and you cure the violation prior to 30 days after 427 | your receipt of the notice. 428 | 429 | Termination of your rights under this section does not terminate the 430 | licenses of parties who have received copies or rights from you under 431 | this License. If your rights have been terminated and not permanently 432 | reinstated, you do not qualify to receive new licenses for the same 433 | material under section 10. 434 | 435 | 9. Acceptance Not Required for Having Copies. 436 | 437 | You are not required to accept this License in order to receive or 438 | run a copy of the Program. Ancillary propagation of a covered work 439 | occurring solely as a consequence of using peer-to-peer transmission 440 | to receive a copy likewise does not require acceptance. However, 441 | nothing other than this License grants you permission to propagate or 442 | modify any covered work. These actions infringe copyright if you do 443 | not accept this License. Therefore, by modifying or propagating a 444 | covered work, you indicate your acceptance of this License to do so. 445 | 446 | 10. Automatic Licensing of Downstream Recipients. 447 | 448 | Each time you convey a covered work, the recipient automatically 449 | receives a license from the original licensors, to run, modify and 450 | propagate that work, subject to this License. You are not responsible 451 | for enforcing compliance by third parties with this License. 452 | 453 | An "entity transaction" is a transaction transferring control of an 454 | organization, or substantially all assets of one, or subdividing an 455 | organization, or merging organizations. If propagation of a covered 456 | work results from an entity transaction, each party to that 457 | transaction who receives a copy of the work also receives whatever 458 | licenses to the work the party's predecessor in interest had or could 459 | give under the previous paragraph, plus a right to possession of the 460 | Corresponding Source of the work from the predecessor in interest, if 461 | the predecessor has it or can get it with reasonable efforts. 462 | 463 | You may not impose any further restrictions on the exercise of the 464 | rights granted or affirmed under this License. For example, you may 465 | not impose a license fee, royalty, or other charge for exercise of 466 | rights granted under this License, and you may not initiate litigation 467 | (including a cross-claim or counterclaim in a lawsuit) alleging that 468 | any patent claim is infringed by making, using, selling, offering for 469 | sale, or importing the Program or any portion of it. 470 | 471 | 11. Patents. 472 | 473 | A "contributor" is a copyright holder who authorizes use under this 474 | License of the Program or a work on which the Program is based. The 475 | work thus licensed is called the contributor's "contributor version". 476 | 477 | A contributor's "essential patent claims" are all patent claims 478 | owned or controlled by the contributor, whether already acquired or 479 | hereafter acquired, that would be infringed by some manner, permitted 480 | by this License, of making, using, or selling its contributor version, 481 | but do not include claims that would be infringed only as a 482 | consequence of further modification of the contributor version. For 483 | purposes of this definition, "control" includes the right to grant 484 | patent sublicenses in a manner consistent with the requirements of 485 | this License. 486 | 487 | Each contributor grants you a non-exclusive, worldwide, royalty-free 488 | patent license under the contributor's essential patent claims, to 489 | make, use, sell, offer for sale, import and otherwise run, modify and 490 | propagate the contents of its contributor version. 491 | 492 | In the following three paragraphs, a "patent license" is any express 493 | agreement or commitment, however denominated, not to enforce a patent 494 | (such as an express permission to practice a patent or covenant not to 495 | sue for patent infringement). To "grant" such a patent license to a 496 | party means to make such an agreement or commitment not to enforce a 497 | patent against the party. 498 | 499 | If you convey a covered work, knowingly relying on a patent license, 500 | and the Corresponding Source of the work is not available for anyone 501 | to copy, free of charge and under the terms of this License, through a 502 | publicly available network server or other readily accessible means, 503 | then you must either (1) cause the Corresponding Source to be so 504 | available, or (2) arrange to deprive yourself of the benefit of the 505 | patent license for this particular work, or (3) arrange, in a manner 506 | consistent with the requirements of this License, to extend the patent 507 | license to downstream recipients. "Knowingly relying" means you have 508 | actual knowledge that, but for the patent license, your conveying the 509 | covered work in a country, or your recipient's use of the covered work 510 | in a country, would infringe one or more identifiable patents in that 511 | country that you have reason to believe are valid. 512 | 513 | If, pursuant to or in connection with a single transaction or 514 | arrangement, you convey, or propagate by procuring conveyance of, a 515 | covered work, and grant a patent license to some of the parties 516 | receiving the covered work authorizing them to use, propagate, modify 517 | or convey a specific copy of the covered work, then the patent license 518 | you grant is automatically extended to all recipients of the covered 519 | work and works based on it. 520 | 521 | A patent license is "discriminatory" if it does not include within 522 | the scope of its coverage, prohibits the exercise of, or is 523 | conditioned on the non-exercise of one or more of the rights that are 524 | specifically granted under this License. You may not convey a covered 525 | work if you are a party to an arrangement with a third party that is 526 | in the business of distributing software, under which you make payment 527 | to the third party based on the extent of your activity of conveying 528 | the work, and under which the third party grants, to any of the 529 | parties who would receive the covered work from you, a discriminatory 530 | patent license (a) in connection with copies of the covered work 531 | conveyed by you (or copies made from those copies), or (b) primarily 532 | for and in connection with specific products or compilations that 533 | contain the covered work, unless you entered into that arrangement, 534 | or that patent license was granted, prior to 28 March 2007. 535 | 536 | Nothing in this License shall be construed as excluding or limiting 537 | any implied license or other defenses to infringement that may 538 | otherwise be available to you under applicable patent law. 539 | 540 | 12. No Surrender of Others' Freedom. 541 | 542 | If conditions are imposed on you (whether by court order, agreement or 543 | otherwise) that contradict the conditions of this License, they do not 544 | excuse you from the conditions of this License. If you cannot convey a 545 | covered work so as to satisfy simultaneously your obligations under this 546 | License and any other pertinent obligations, then as a consequence you may 547 | not convey it at all. For example, if you agree to terms that obligate you 548 | to collect a royalty for further conveying from those to whom you convey 549 | the Program, the only way you could satisfy both those terms and this 550 | License would be to refrain entirely from conveying the Program. 551 | 552 | 13. Use with the GNU Affero General Public License. 553 | 554 | Notwithstanding any other provision of this License, you have 555 | permission to link or combine any covered work with a work licensed 556 | under version 3 of the GNU Affero General Public License into a single 557 | combined work, and to convey the resulting work. The terms of this 558 | License will continue to apply to the part which is the covered work, 559 | but the special requirements of the GNU Affero General Public License, 560 | section 13, concerning interaction through a network will apply to the 561 | combination as such. 562 | 563 | 14. Revised Versions of this License. 564 | 565 | The Free Software Foundation may publish revised and/or new versions of 566 | the GNU General Public License from time to time. Such new versions will 567 | be similar in spirit to the present version, but may differ in detail to 568 | address new problems or concerns. 569 | 570 | Each version is given a distinguishing version number. If the 571 | Program specifies that a certain numbered version of the GNU General 572 | Public License "or any later version" applies to it, you have the 573 | option of following the terms and conditions either of that numbered 574 | version or of any later version published by the Free Software 575 | Foundation. If the Program does not specify a version number of the 576 | GNU General Public License, you may choose any version ever published 577 | by the Free Software Foundation. 578 | 579 | If the Program specifies that a proxy can decide which future 580 | versions of the GNU General Public License can be used, that proxy's 581 | public statement of acceptance of a version permanently authorizes you 582 | to choose that version for the Program. 583 | 584 | Later license versions may give you additional or different 585 | permissions. However, no additional obligations are imposed on any 586 | author or copyright holder as a result of your choosing to follow a 587 | later version. 588 | 589 | 15. Disclaimer of Warranty. 590 | 591 | THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY 592 | APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT 593 | HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY 594 | OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, 595 | THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 596 | PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM 597 | IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF 598 | ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 599 | 600 | 16. Limitation of Liability. 601 | 602 | IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING 603 | WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS 604 | THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY 605 | GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE 606 | USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF 607 | DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD 608 | PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), 609 | EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF 610 | SUCH DAMAGES. 611 | 612 | 17. Interpretation of Sections 15 and 16. 613 | 614 | If the disclaimer of warranty and limitation of liability provided 615 | above cannot be given local legal effect according to their terms, 616 | reviewing courts shall apply local law that most closely approximates 617 | an absolute waiver of all civil liability in connection with the 618 | Program, unless a warranty or assumption of liability accompanies a 619 | copy of the Program in return for a fee. 620 | 621 | END OF TERMS AND CONDITIONS 622 | 623 | How to Apply These Terms to Your New Programs 624 | 625 | If you develop a new program, and you want it to be of the greatest 626 | possible use to the public, the best way to achieve this is to make it 627 | free software which everyone can redistribute and change under these terms. 628 | 629 | To do so, attach the following notices to the program. It is safest 630 | to attach them to the start of each source file to most effectively 631 | state the exclusion of warranty; and each file should have at least 632 | the "copyright" line and a pointer to where the full notice is found. 633 | 634 | 635 | Copyright (C) 636 | 637 | This program is free software: you can redistribute it and/or modify 638 | it under the terms of the GNU General Public License as published by 639 | the Free Software Foundation, either version 3 of the License, or 640 | (at your option) any later version. 641 | 642 | This program is distributed in the hope that it will be useful, 643 | but WITHOUT ANY WARRANTY; without even the implied warranty of 644 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 645 | GNU General Public License for more details. 646 | 647 | You should have received a copy of the GNU General Public License 648 | along with this program. If not, see . 649 | 650 | Also add information on how to contact you by electronic and paper mail. 651 | 652 | If the program does terminal interaction, make it output a short 653 | notice like this when it starts in an interactive mode: 654 | 655 | Copyright (C) 656 | This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'. 657 | This is free software, and you are welcome to redistribute it 658 | under certain conditions; type `show c' for details. 659 | 660 | The hypothetical commands `show w' and `show c' should show the appropriate 661 | parts of the General Public License. Of course, your program's commands 662 | might be different; for a GUI interface, you would use an "about box". 663 | 664 | You should also get your employer (if you work as a programmer) or school, 665 | if any, to sign a "copyright disclaimer" for the program, if necessary. 666 | For more information on this, and how to apply and follow the GNU GPL, see 667 | . 668 | 669 | The GNU General Public License does not permit incorporating your program 670 | into proprietary programs. If your program is a subroutine library, you 671 | may consider it more useful to permit linking proprietary applications with 672 | the library. If this is what you want to do, use the GNU Lesser General 673 | Public License instead of this License. But first, please read 674 | . 675 | -------------------------------------------------------------------------------- /Model Interpretation.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "markdown", 5 | "metadata": {}, 6 | "source": [ 7 | "### Retrieving information from the predictor insight table\n", 8 | "- The predictor insight graph table contains all the information needed to construct the predictor insight graph. For each value the predictor takes, it has the number of observations with this value and the target incidence within this group. The predictor insight graph table of the predictor Country is loaded as a pandas object pig_table" 9 | ] 10 | }, 11 | { 12 | "cell_type": "code", 13 | "execution_count": 1, 14 | "metadata": {}, 15 | "outputs": [], 16 | "source": [ 17 | "import pandas as pd" 18 | ] 19 | }, 20 | { 21 | "cell_type": "code", 22 | "execution_count": 3, 23 | "metadata": {}, 24 | "outputs": [ 25 | { 26 | "data": { 27 | "text/html": [ 28 | "
\n", 29 | "\n", 42 | "\n", 43 | " \n", 44 | " \n", 45 | " \n", 46 | " \n", 47 | " \n", 48 | " \n", 49 | " \n", 50 | " \n", 51 | " \n", 52 | " \n", 53 | " \n", 54 | " \n", 55 | " \n", 56 | " \n", 57 | " \n", 58 | " \n", 59 | " \n", 60 | " \n", 61 | " \n", 62 | " \n", 63 | " \n", 64 | " \n", 65 | " \n", 66 | " \n", 67 | " \n", 68 | " \n", 69 | " \n", 70 | " \n", 71 | "
CountrySizeIncidence
0India498490.05
1UK100570.05
2USA400940.05
\n", 72 | "
" 73 | ], 74 | "text/plain": [ 75 | " Country Size Incidence\n", 76 | "0 India 49849 0.05\n", 77 | "1 UK 10057 0.05\n", 78 | "2 USA 40094 0.05" 79 | ] 80 | }, 81 | "execution_count": 3, 82 | "metadata": {}, 83 | "output_type": "execute_result" 84 | } 85 | ], 86 | "source": [ 87 | "pig_table = pd.DataFrame({'Country':['India','UK','USA'], 'Size':[49849,10057,40094], 'Incidence':[0.05,0.05,0.05]})\n", 88 | "pig_table" 89 | ] 90 | }, 91 | { 92 | "cell_type": "code", 93 | "execution_count": 4, 94 | "metadata": {}, 95 | "outputs": [ 96 | { 97 | "name": "stdout", 98 | "output_type": "stream", 99 | "text": [ 100 | " Country Size Incidence\n", 101 | "0 India 49849 0.05\n", 102 | "1 UK 10057 0.05\n", 103 | "2 USA 40094 0.05\n" 104 | ] 105 | } 106 | ], 107 | "source": [ 108 | "# Inspect the predictor insight graph table of Country\n", 109 | "print(pig_table)" 110 | ] 111 | }, 112 | { 113 | "cell_type": "code", 114 | "execution_count": 5, 115 | "metadata": {}, 116 | "outputs": [ 117 | { 118 | "name": "stdout", 119 | "output_type": "stream", 120 | "text": [ 121 | "1 10057\n", 122 | "Name: Size, dtype: int64\n" 123 | ] 124 | } 125 | ], 126 | "source": [ 127 | "# Print the number of UK donors\n", 128 | "print(pig_table[\"Size\"][pig_table[\"Country\"]==\"UK\"])" 129 | ] 130 | }, 131 | { 132 | "cell_type": "code", 133 | "execution_count": 6, 134 | "metadata": {}, 135 | "outputs": [ 136 | { 137 | "name": "stdout", 138 | "output_type": "stream", 139 | "text": [ 140 | "2 0.05\n", 141 | "Name: Incidence, dtype: float64\n", 142 | "0 0.05\n", 143 | "Name: Incidence, dtype: float64\n" 144 | ] 145 | } 146 | ], 147 | "source": [ 148 | "# Check the target incidence of USA and India donors\n", 149 | "print(pig_table[\"Incidence\"][pig_table[\"Country\"]==\"USA\"])\n", 150 | "print(pig_table[\"Incidence\"][pig_table[\"Country\"]==\"India\"])" 151 | ] 152 | }, 153 | { 154 | "cell_type": "markdown", 155 | "metadata": {}, 156 | "source": [ 157 | "### The target incidence of USA and India donors is the same, indicating that country is not a good variable to predict donations." 158 | ] 159 | }, 160 | { 161 | "cell_type": "markdown", 162 | "metadata": {}, 163 | "source": [ 164 | "#### Discretization of a certain variable\n", 165 | "- In order to make predictor insight graphs for continuous variables, we first need to discretize them. In Python, we can discretize pandas columns using the `qcut` method.\n", 166 | "- To check whether the variable was nicely discretized, we can verify that the bins have equal size using the groupby method:\n", 167 | "`print(basetable.groupby(\"discretized_variable\").size()`" 168 | ] 169 | }, 170 | { 171 | "cell_type": "code", 172 | "execution_count": null, 173 | "metadata": {}, 174 | "outputs": [], 175 | "source": [ 176 | "# Discretize the variable time_since_last_donation in 10 bins\n", 177 | "basetable[\"bins_recency\"] = pd.qcut(basetable['time_since_last_donation'],10)\n", 178 | "\n", 179 | "# Print the group sizes of the discretized variable\n", 180 | "print(basetable.groupby(\"bins_recency\").size())" 181 | ] 182 | }, 183 | { 184 | "cell_type": "markdown", 185 | "metadata": {}, 186 | "source": [ 187 | "### Discretizing all variables\n", 188 | "- Instead of discretizing the continuous variables one by one, it is easier to discretize them automatically. \n", 189 | "- Only variables that are continuous should be discretized. You can verify whether variables should be discretized by checking whether they have more than a predefined number of different values.\n", 190 | "- Only variables that are continuous should be discretized. we can verify whether variables should be discretized by checking whether they have more than a predefined number of different values.\n", 191 | "\n", 192 | "\n", 193 | "\n", 194 | "- Make a list variables containing all the column names of the basetable.\n", 195 | "- Create a loop that checks all the variables in the list variables.\n", 196 | "- Complete the ifstatement such that only variables with more than 5 different values are discretized.\n", 197 | "- Group the continuous variables in 10 bins using the qcut method.\n" 198 | ] 199 | }, 200 | { 201 | "cell_type": "code", 202 | "execution_count": null, 203 | "metadata": {}, 204 | "outputs": [], 205 | "source": [ 206 | "# Print the columns in the original basetable\n", 207 | "print(basetable.columns)\n", 208 | "\n", 209 | "# Get all the variable names except \"target\"\n", 210 | "variables = list(basetable.columns)\n", 211 | "variables.remove(\"target\")\n", 212 | "\n", 213 | "# Loop through all the variables and discretize in 10 bins if there are more than 5 different values\n", 214 | "for variable in variables:\n", 215 | " if len(basetable.groupby(variable))>5:\n", 216 | " new_variable = \"disc_\" + variable\n", 217 | " basetable[new_variable] = pd.qcut(basetable[variable], 10)\n", 218 | " \n", 219 | "# Print the columns in the new basetable\n", 220 | "print(basetable.columns)" 221 | ] 222 | }, 223 | { 224 | "cell_type": "markdown", 225 | "metadata": {}, 226 | "source": [ 227 | "```python\n", 228 | "Index(['target', 'gender_F', 'gender_M', 'income_average', 'income_low',\n", 229 | " 'income_high', 'country_USA', 'country_India', 'country_UK', 'age',\n", 230 | " 'time_since_last_gift', 'time_since_first_gift', 'max_gift', 'min_gift',\n", 231 | " 'mean_gift', 'median_gift'],\n", 232 | " dtype='object')\n", 233 | "Index(['target', 'gender_F', 'gender_M', 'income_average', 'income_low',\n", 234 | " 'income_high', 'country_USA', 'country_India', 'country_UK', 'age',\n", 235 | " 'time_since_last_gift', 'time_since_first_gift', 'max_gift', 'min_gift',\n", 236 | " 'mean_gift', 'median_gift', 'disc_age', 'disc_time_since_last_gift',\n", 237 | " 'disc_time_since_first_gift', 'disc_max_gift', 'disc_min_gift',\n", 238 | " 'disc_mean_gift', 'disc_median_gift'],\n", 239 | " dtype='object')\n", 240 | "```" 241 | ] 242 | }, 243 | { 244 | "cell_type": "markdown", 245 | "metadata": {}, 246 | "source": [ 247 | "### Making clean cuts\n", 248 | "- The `qcut` method divides the variable in n_bins equal bins. In some cases, however, it is nice to choose our own bins. The method cut in python allows us to choose our own bins.\n", 249 | "\n", 250 | "- Discretize the variable number_gift in three bins with borders 0 and 5, 5 and 10, 10 and 20 and assign this variable to a new column called `disc_number_gift`.\n", 251 | "- Count the number of observations in each group.\n" 252 | ] 253 | }, 254 | { 255 | "cell_type": "code", 256 | "execution_count": null, 257 | "metadata": {}, 258 | "outputs": [], 259 | "source": [ 260 | "# Discretize the variable \n", 261 | "basetable[\"disc_number_gift\"] = pd.cut(basetable['number_gift'],[0,5,10, 20])\n", 262 | "\n", 263 | "# Count the number of observations per group\n", 264 | "print(basetable.groupby(\"disc_number_gift\").size())" 265 | ] 266 | }, 267 | { 268 | "cell_type": "markdown", 269 | "metadata": {}, 270 | "source": [ 271 | "```python\n", 272 | "disc_number_gift\n", 273 | "(0, 5) 55063\n", 274 | "(5, 10) 41120\n", 275 | "(10, 20) 3817\n", 276 | "dtype: int64\n", 277 | "```\n", 278 | "\n", 279 | "- Notice that the bins aren't approximately equally sized anymore." 280 | ] 281 | } 282 | ], 283 | "metadata": { 284 | "kernelspec": { 285 | "display_name": "Python 3", 286 | "language": "python", 287 | "name": "python3" 288 | }, 289 | "language_info": { 290 | "codemirror_mode": { 291 | "name": "ipython", 292 | "version": 3 293 | }, 294 | "file_extension": ".py", 295 | "mimetype": "text/x-python", 296 | "name": "python", 297 | "nbconvert_exporter": "python", 298 | "pygments_lexer": "ipython3", 299 | "version": "3.7.3" 300 | } 301 | }, 302 | "nbformat": 4, 303 | "nbformat_minor": 2 304 | } 305 | -------------------------------------------------------------------------------- /Model_Performance.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "markdown", 5 | "metadata": {}, 6 | "source": [ 7 | "# Constructing the cumulative gains curve\n", 8 | "- The cumulative gains curve is an evaluation curve that assesses the performance of your model. It shows the percentage of targets reached when considering a certain percentage of your population with the highest probability to be target according to your model. \n", 9 | "- To construct this curve, we can use the `.plot_cumulative_gain()` method in the `scikitplot` module and the `matplotlib.pyplot` module. As for each model evaluation metric or curve, we need the true target values on the one hand and the predictions on the other hand to construct the cumulative gains curve." 10 | ] 11 | }, 12 | { 13 | "cell_type": "code", 14 | "execution_count": 12, 15 | "metadata": {}, 16 | "outputs": [], 17 | "source": [ 18 | "import pandas as pd\n", 19 | "from sklearn.linear_model import LogisticRegression\n", 20 | "from sklearn.model_selection import train_test_split\n", 21 | "\n", 22 | "# Import the matplotlib.pyplot module \n", 23 | "import matplotlib.pyplot as plt\n", 24 | "\n", 25 | "# Import the scikitplot module\n", 26 | "import scikitplot as skplt" 27 | ] 28 | }, 29 | { 30 | "cell_type": "code", 31 | "execution_count": 2, 32 | "metadata": {}, 33 | "outputs": [], 34 | "source": [ 35 | "basetable = pd.read_csv('basetable_ex2_4.csv')" 36 | ] 37 | }, 38 | { 39 | "cell_type": "code", 40 | "execution_count": 15, 41 | "metadata": {}, 42 | "outputs": [ 43 | { 44 | "name": "stderr", 45 | "output_type": "stream", 46 | "text": [ 47 | "C:\\Users\\Shubham\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n", 48 | " FutureWarning)\n", 49 | "C:\\Users\\Shubham\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:761: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", 50 | " y = column_or_1d(y, warn=True)\n" 51 | ] 52 | } 53 | ], 54 | "source": [ 55 | "X = basetable[[\"age\", \"gender_F\", \"time_since_last_gift\"]]\n", 56 | "y = basetable[[\"target\"]]\n", 57 | "\n", 58 | "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=123)\n", 59 | "logreg = LogisticRegression()\n", 60 | "logreg.fit(X_train, y_train)\n", 61 | "pred = logreg.predict_proba(X_test)" 62 | ] 63 | }, 64 | { 65 | "cell_type": "code", 66 | "execution_count": 17, 67 | "metadata": {}, 68 | "outputs": [ 69 | { 70 | "name": "stderr", 71 | "output_type": "stream", 72 | "text": [ 73 | "C:\\Users\\Shubham\\Anaconda3\\lib\\site-packages\\matplotlib\\cbook\\__init__.py:424: MatplotlibDeprecationWarning: \n", 74 | "Passing one of 'on', 'true', 'off', 'false' as a boolean is deprecated; use an actual boolean (True/False) instead.\n", 75 | " warn_deprecated(\"2.2\", \"Passing one of 'on', 'true', 'off', 'false' as a \"\n" 76 | ] 77 | }, 78 | { 79 | "data": { 80 | "image/png": "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\n", 81 | "text/plain": [ 82 | "
" 83 | ] 84 | }, 85 | "metadata": { 86 | "needs_background": "light" 87 | }, 88 | "output_type": "display_data" 89 | } 90 | ], 91 | "source": [ 92 | "# Plot the cumulative gains graph\n", 93 | "skplt.metrics.plot_cumulative_gain(y_test, pred)\n", 94 | "plt.show()" 95 | ] 96 | }, 97 | { 98 | "cell_type": "markdown", 99 | "metadata": {}, 100 | "source": [ 101 | "## A random model\n", 102 | "- We will reconstruct the cumulative gains curve's baseline, that is, the cumulative gains curve of a random model.\n", 103 | "- To do so, we need to construct random predictions. The `plot_cumulative_gain` method requires two values for these predictions: one for the target to be 0 and one for the target to be 1. These values should sum to one, so a valid list of predictions could for instance be [(0.02,0.98),(0.27,0.73),...,(0.09,0.91)].\n", 104 | "\n", 105 | "- In Python, we can generate a random value between values a and b as follows:\n", 106 | "\n", 107 | "```python \n", 108 | "import random\n", 109 | "random_value = random.uniform(a,b)\n", 110 | "```" 111 | ] 112 | }, 113 | { 114 | "cell_type": "markdown", 115 | "metadata": {}, 116 | "source": [ 117 | "- Construct a list random_predictions that contains random numbers between 0 and 1.\n", 118 | "- Adjust the list random_predictions such that it contains tuples (r,a) with r the original value of the list and a such that r+a=1\n", 119 | ".\n", 120 | "- The true values of the target are in targets_test. Show the cumulative gains graph of your random model." 121 | ] 122 | }, 123 | { 124 | "cell_type": "code", 125 | "execution_count": 23, 126 | "metadata": {}, 127 | "outputs": [], 128 | "source": [ 129 | "import random\n", 130 | "import matplotlib.pyplot as plt\n", 131 | "import scikitplot as skplt" 132 | ] 133 | }, 134 | { 135 | "cell_type": "code", 136 | "execution_count": 22, 137 | "metadata": {}, 138 | "outputs": [ 139 | { 140 | "data": { 141 | "image/png": "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\n", 142 | "text/plain": [ 143 | "
" 144 | ] 145 | }, 146 | "metadata": { 147 | "needs_background": "light" 148 | }, 149 | "output_type": "display_data" 150 | } 151 | ], 152 | "source": [ 153 | "# Generate random predictions\n", 154 | "random_predictions = [random.uniform(0,1) for _ in range(len(y_test))]\n", 155 | "\n", 156 | "# Adjust random predictions\n", 157 | "random_predictions = [(r, 1- r) for r in random_predictions]\n", 158 | "\n", 159 | "# Plot the cumulative gains graph\n", 160 | "skplt.metrics.plot_cumulative_gain(y_test, random_predictions)\n", 161 | "plt.show()" 162 | ] 163 | }, 164 | { 165 | "cell_type": "markdown", 166 | "metadata": {}, 167 | "source": [ 168 | "#### We can observe that the cumulative gains curve of a random model aligns with the baseline." 169 | ] 170 | }, 171 | { 172 | "cell_type": "markdown", 173 | "metadata": {}, 174 | "source": [] 175 | }, 176 | { 177 | "cell_type": "markdown", 178 | "metadata": {}, 179 | "source": [ 180 | "### Constructing the lift curve\n", 181 | "- The lift curve is an evaluation curve that assesses the performance of our model. It shows how many times more than average the model reaches targets.\n", 182 | "- To construct this curve, we can use the `plot_lift_curve` method in the `scikitplot` module and the `matplotlib.pyplot` module. As for each model evaluation metric or curve, we need the true target values on the one hand and the predictions on the other hand to construct the cumulative gains curve." 183 | ] 184 | }, 185 | { 186 | "cell_type": "code", 187 | "execution_count": 24, 188 | "metadata": {}, 189 | "outputs": [ 190 | { 191 | "data": { 192 | "image/png": "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\n", 193 | "text/plain": [ 194 | "
" 195 | ] 196 | }, 197 | "metadata": { 198 | "needs_background": "light" 199 | }, 200 | "output_type": "display_data" 201 | } 202 | ], 203 | "source": [ 204 | "# Plot the lift curve\n", 205 | "skplt.metrics.plot_lift_curve(y_test, random_predictions)\n", 206 | "plt.show()" 207 | ] 208 | }, 209 | { 210 | "cell_type": "markdown", 211 | "metadata": {}, 212 | "source": [ 213 | "#### A perfect model\n", 214 | "- reconstruct the lift curve of a perfect model. To do so, you need to construct perfect predictions. `plot_lift_curve` method requires two values for the predictions argument: the first argument for the target to be 0 and the second one for the target to be 1." 215 | ] 216 | }, 217 | { 218 | "cell_type": "code", 219 | "execution_count": 28, 220 | "metadata": {}, 221 | "outputs": [ 222 | { 223 | "name": "stderr", 224 | "output_type": "stream", 225 | "text": [ 226 | "C:\\Users\\Shubham\\Anaconda3\\lib\\site-packages\\matplotlib\\cbook\\__init__.py:424: MatplotlibDeprecationWarning: \n", 227 | "Passing one of 'on', 'true', 'off', 'false' as a boolean is deprecated; use an actual boolean (True/False) instead.\n", 228 | " warn_deprecated(\"2.2\", \"Passing one of 'on', 'true', 'off', 'false' as a \"\n" 229 | ] 230 | }, 231 | { 232 | "data": { 233 | "image/png": "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\n", 234 | "text/plain": [ 235 | "
" 236 | ] 237 | }, 238 | "metadata": { 239 | "needs_background": "light" 240 | }, 241 | "output_type": "display_data" 242 | } 243 | ], 244 | "source": [ 245 | "# Generate perfect predictions\n", 246 | "perfect_predictions = [(1-target[0] , target[0]) for target in random_predictions]\n", 247 | "\n", 248 | "# Plot the lift curve\n", 249 | "skplt.metrics.plot_lift_curve(y_test, perfect_predictions)\n", 250 | "plt.show()" 251 | ] 252 | } 253 | ], 254 | "metadata": { 255 | "kernelspec": { 256 | "display_name": "Python 3", 257 | "language": "python", 258 | "name": "python3" 259 | }, 260 | "language_info": { 261 | "codemirror_mode": { 262 | "name": "ipython", 263 | "version": 3 264 | }, 265 | "file_extension": ".py", 266 | "mimetype": "text/x-python", 267 | "name": "python", 268 | "nbconvert_exporter": "python", 269 | "pygments_lexer": "ipython3", 270 | "version": "3.7.3" 271 | } 272 | }, 273 | "nbformat": 4, 274 | "nbformat_minor": 2 275 | } 276 | -------------------------------------------------------------------------------- /Predictive analytics.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "code", 5 | "execution_count": 45, 6 | "metadata": {}, 7 | "outputs": [], 8 | "source": [ 9 | "import pandas as pd\n", 10 | "from sklearn import linear_model\n", 11 | "from sklearn.metrics import roc_auc_score\n", 12 | "import matplotlib.pyplot as plt" 13 | ] 14 | }, 15 | { 16 | "cell_type": "code", 17 | "execution_count": 33, 18 | "metadata": {}, 19 | "outputs": [], 20 | "source": [ 21 | "basetable = pd.read_csv('basetable_ex2_4.csv')" 22 | ] 23 | }, 24 | { 25 | "cell_type": "code", 26 | "execution_count": 3, 27 | "metadata": {}, 28 | "outputs": [ 29 | { 30 | "data": { 31 | "text/html": [ 32 | "
\n", 33 | "\n", 46 | "\n", 47 | " \n", 48 | " \n", 49 | " \n", 50 | " \n", 51 | " \n", 52 | " \n", 53 | " \n", 54 | " \n", 55 | " \n", 56 | " \n", 57 | " \n", 58 | " \n", 59 | " \n", 60 | " \n", 61 | " \n", 62 | " \n", 63 | " \n", 64 | " \n", 65 | " \n", 66 | " \n", 67 | " \n", 68 | " \n", 69 | " \n", 70 | " \n", 71 | " \n", 72 | " \n", 73 | " \n", 74 | " \n", 75 | " \n", 76 | " \n", 77 | " \n", 78 | " \n", 79 | " \n", 80 | " \n", 81 | " \n", 82 | " \n", 83 | " \n", 84 | " \n", 85 | " \n", 86 | " \n", 87 | " \n", 88 | " \n", 89 | " \n", 90 | " \n", 91 | " \n", 92 | " \n", 93 | " \n", 94 | " \n", 95 | " \n", 96 | " \n", 97 | " \n", 98 | " \n", 99 | " \n", 100 | " \n", 101 | " \n", 102 | " \n", 103 | " \n", 104 | " \n", 105 | " \n", 106 | " \n", 107 | " \n", 108 | " \n", 109 | " \n", 110 | " \n", 111 | " \n", 112 | " \n", 113 | " \n", 114 | " \n", 115 | " \n", 116 | " \n", 117 | " \n", 118 | " \n", 119 | " \n", 120 | " \n", 121 | " \n", 122 | " \n", 123 | " \n", 124 | " \n", 125 | " \n", 126 | " \n", 127 | " \n", 128 | " \n", 129 | " \n", 130 | " \n", 131 | " \n", 132 | " \n", 133 | " \n", 134 | " \n", 135 | " \n", 136 | " \n", 137 | " \n", 138 | " \n", 139 | " \n", 140 | " \n", 141 | " \n", 142 | " \n", 143 | " \n", 144 | " \n", 145 | " \n", 146 | " \n", 147 | " \n", 148 | " \n", 149 | " \n", 150 | " \n", 151 | " \n", 152 | " \n", 153 | "
targetgender_Fincome_highincome_lowcountry_USAcountry_Indiacountry_UKagetime_since_last_gifttime_since_first_giftmax_giftmin_giftmean_giftnumber_gift
00101010655302265166.087.0116.007
101000107171571590.090.090.001
20100010281501806125.074.096.009
30101100527252274117.097.0104.254
401101008280580580.080.080.001
\n", 154 | "
" 155 | ], 156 | "text/plain": [ 157 | " target gender_F income_high income_low country_USA country_India \\\n", 158 | "0 0 1 0 1 0 1 \n", 159 | "1 0 1 0 0 0 1 \n", 160 | "2 0 1 0 0 0 1 \n", 161 | "3 0 1 0 1 1 0 \n", 162 | "4 0 1 1 0 1 0 \n", 163 | "\n", 164 | " country_UK age time_since_last_gift time_since_first_gift max_gift \\\n", 165 | "0 0 65 530 2265 166.0 \n", 166 | "1 0 71 715 715 90.0 \n", 167 | "2 0 28 150 1806 125.0 \n", 168 | "3 0 52 725 2274 117.0 \n", 169 | "4 0 82 805 805 80.0 \n", 170 | "\n", 171 | " min_gift mean_gift number_gift \n", 172 | "0 87.0 116.00 7 \n", 173 | "1 90.0 90.00 1 \n", 174 | "2 74.0 96.00 9 \n", 175 | "3 97.0 104.25 4 \n", 176 | "4 80.0 80.00 1 " 177 | ] 178 | }, 179 | "execution_count": 3, 180 | "metadata": {}, 181 | "output_type": "execute_result" 182 | } 183 | ], 184 | "source": [ 185 | "basetable.head()" 186 | ] 187 | }, 188 | { 189 | "cell_type": "code", 190 | "execution_count": 6, 191 | "metadata": {}, 192 | "outputs": [ 193 | { 194 | "name": "stderr", 195 | "output_type": "stream", 196 | "text": [ 197 | "C:\\Users\\Shubham\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n", 198 | " FutureWarning)\n", 199 | "C:\\Users\\Shubham\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:761: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", 200 | " y = column_or_1d(y, warn=True)\n" 201 | ] 202 | }, 203 | { 204 | "data": { 205 | "text/plain": [ 206 | "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n", 207 | " intercept_scaling=1, max_iter=100, multi_class='warn',\n", 208 | " n_jobs=None, penalty='l2', random_state=None, solver='warn',\n", 209 | " tol=0.0001, verbose=0, warm_start=False)" 210 | ] 211 | }, 212 | "execution_count": 6, 213 | "metadata": {}, 214 | "output_type": "execute_result" 215 | } 216 | ], 217 | "source": [ 218 | "X = basetable[[\"age\", \"gender_F\", \"time_since_last_gift\"]]\n", 219 | "y = basetable[[\"target\"]]\n", 220 | "logreg = linear_model.LogisticRegression()\n", 221 | "logreg.fit(X, y)" 222 | ] 223 | }, 224 | { 225 | "cell_type": "code", 226 | "execution_count": 5, 227 | "metadata": {}, 228 | "outputs": [ 229 | { 230 | "name": "stdout", 231 | "output_type": "stream", 232 | "text": [ 233 | "[0 0 0 0 0]\n" 234 | ] 235 | } 236 | ], 237 | "source": [ 238 | "# Create a dataframe new_data from current_data that has only the relevant predictors \n", 239 | "new_data = X[[\"age\", \"gender_F\", \"time_since_last_gift\"]]\n", 240 | "\n", 241 | "# Make a prediction for each observation in new_data and assign it to predictions\n", 242 | "predictions = logreg.predict(new_data)\n", 243 | "print(predictions[0:5])" 244 | ] 245 | }, 246 | { 247 | "cell_type": "markdown", 248 | "metadata": {}, 249 | "source": [ 250 | "### Calculating AUC\n", 251 | "- The AUC value assesses how well a model can order observations from low probability to be target to high probability to be target. In Python, the roc_auc_score function can be used to calculate the AUC of the model. It takes the true values of the target and the predictions as arguments. " 252 | ] 253 | }, 254 | { 255 | "cell_type": "code", 256 | "execution_count": 10, 257 | "metadata": {}, 258 | "outputs": [ 259 | { 260 | "name": "stdout", 261 | "output_type": "stream", 262 | "text": [ 263 | "0.63\n" 264 | ] 265 | } 266 | ], 267 | "source": [ 268 | "# Make predictions\n", 269 | "predictions = logreg.predict_proba(X)\n", 270 | "predictions_target = predictions[:,-1]\n", 271 | "\n", 272 | "# Calculate the AUC value\n", 273 | "auc = roc_auc_score(y, predictions_target)\n", 274 | "print(round(auc,2))" 275 | ] 276 | }, 277 | { 278 | "cell_type": "markdown", 279 | "metadata": {}, 280 | "source": [ 281 | "### Using different set of variables to calculate AUC score\n", 282 | "- Adding more variables and therefore more complexity to our logistic regression model does not automatically result in more accurate models. Here we can verify whether adding 3 variables to a model leads to a more accurate model." 283 | ] 284 | }, 285 | { 286 | "cell_type": "code", 287 | "execution_count": 7, 288 | "metadata": {}, 289 | "outputs": [ 290 | { 291 | "name": "stdout", 292 | "output_type": "stream", 293 | "text": [ 294 | "0.68\n", 295 | "0.69\n" 296 | ] 297 | }, 298 | { 299 | "name": "stderr", 300 | "output_type": "stream", 301 | "text": [ 302 | "C:\\Users\\Shubham\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n", 303 | " FutureWarning)\n", 304 | "C:\\Users\\Shubham\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:761: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", 305 | " y = column_or_1d(y, warn=True)\n", 306 | "C:\\Users\\Shubham\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n", 307 | " FutureWarning)\n", 308 | "C:\\Users\\Shubham\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\validation.py:761: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n", 309 | " y = column_or_1d(y, warn=True)\n" 310 | ] 311 | } 312 | ], 313 | "source": [ 314 | "# Create appropriate dataframes\n", 315 | "variables_1 = ['mean_gift', 'income_low']\n", 316 | "variables_2 = ['mean_gift', 'income_low', 'gender_F', 'country_India', 'age']\n", 317 | "\n", 318 | "X_1 = basetable[variables_1]\n", 319 | "X_2 = basetable[variables_2]\n", 320 | "y = basetable[[\"target\"]]\n", 321 | "\n", 322 | "# Create the logistic regression model\n", 323 | "logreg = linear_model.LogisticRegression()\n", 324 | "\n", 325 | "# Make predictions using the first set of variables and assign the AUC to auc_1\n", 326 | "logreg.fit(X_1, y)\n", 327 | "predictions_1 = logreg.predict_proba(X_1)[:,1]\n", 328 | "auc_1 = roc_auc_score(y, predictions_1)\n", 329 | "\n", 330 | "# Make predictions using the second set of variables and assign the AUC to auc_2\n", 331 | "logreg.fit(X_2, y)\n", 332 | "predictions_2 = logreg.predict_proba(X_2)[:,1]\n", 333 | "auc_2 = roc_auc_score(y, predictions_2)\n", 334 | "\n", 335 | "# Print auc_1 and auc_2\n", 336 | "print(round(auc_1,2))\n", 337 | "print(round(auc_2,2))" 338 | ] 339 | }, 340 | { 341 | "cell_type": "markdown", 342 | "metadata": {}, 343 | "source": [ 344 | "#### We can see that the model with 5 variables has the same AUC as the model using only 2 variables. Adding more variables doesn't always increase the AUC." 345 | ] 346 | }, 347 | { 348 | "cell_type": "markdown", 349 | "metadata": {}, 350 | "source": [ 351 | "## Forward stepwise variable selection : Intutive way of variable selection\n", 352 | "\n", 353 | "### Selecting the next best variable\n", 354 | "- The forward stepwise variable selection method starts with an empty variable set and proceeds in steps, where in each step the next best variable is added.\n", 355 | "- The **`auc`** function calculates for a given variable set variables the AUC of the model that uses this variable set as predictors. \n", 356 | "- The **`next_best`** function calculates which variable should be added in the next step to the variable list. \n", 357 | "- **Task** : experiment with these functions to better understand their purpose. We will calculate the AUC of a given variable set, calculate which variable should be added next, and verify that this indeed results in an optimal AUC." 358 | ] 359 | }, 360 | { 361 | "cell_type": "code", 362 | "execution_count": 21, 363 | "metadata": {}, 364 | "outputs": [], 365 | "source": [ 366 | "from sklearn import linear_model\n", 367 | "from sklearn.metrics import roc_auc_score\n", 368 | "import warnings\n", 369 | "warnings.filterwarnings(\"ignore\")" 370 | ] 371 | }, 372 | { 373 | "cell_type": "code", 374 | "execution_count": 41, 375 | "metadata": {}, 376 | "outputs": [], 377 | "source": [ 378 | "# function to calculate AUC\n", 379 | "def auc(variables, target, basetable):\n", 380 | " \"\"\"calculates AUC\"\"\"\n", 381 | " X = basetable[variables]\n", 382 | " y = basetable[target]\n", 383 | " \n", 384 | " logreg = linear_model.LogisticRegression()\n", 385 | " logreg.fit(X,y)\n", 386 | " \n", 387 | " predictions = logreg.predict_proba(X)[:,1]\n", 388 | " auc = roc_auc_score(y, predictions)\n", 389 | " return auc\n", 390 | "\n", 391 | "def next_best(current_variable, candidate_variables, target, basetable):\n", 392 | " \"\"\"function looks throughout candidate variables and keeps track of which variable is best and the auc associated with the best variable\"\"\"\n", 393 | " best_auc = -1\n", 394 | " best_variable = None\n", 395 | " \n", 396 | " # for each variable in the candidate variable set calculate the AUC\n", 397 | " # current_variable : variables which are already in the model\n", 398 | " # extend it with the variable with which we need to evaluate\n", 399 | " for v in candidate_variables:\n", 400 | " auc_v = auc(candidate_variables + [v], target, basetable)\n", 401 | " \n", 402 | " # if this AUC is better then the best AUC, change the best AUC and best variable\n", 403 | " if auc_v >= best_auc:\n", 404 | " best_auc = auc_v\n", 405 | " best_variable = v\n", 406 | " return best_variable" 407 | ] 408 | }, 409 | { 410 | "cell_type": "code", 411 | "execution_count": 24, 412 | "metadata": {}, 413 | "outputs": [ 414 | { 415 | "name": "stdout", 416 | "output_type": "stream", 417 | "text": [ 418 | "0.7125\n", 419 | "gender_F\n", 420 | "0.7148\n", 421 | "0.713\n" 422 | ] 423 | } 424 | ], 425 | "source": [ 426 | "# Calculate the AUC of a model that uses \"max_gift\", \"mean_gift\" and \"min_gift\" as predictors\n", 427 | "auc_current = auc([\"max_gift\", \"mean_gift\", \"min_gift\"], [\"target\"], basetable)\n", 428 | "print(round(auc_current,4))\n", 429 | "\n", 430 | "# Calculate which variable among \"age\" and \"gender_F\" should be added to the variables \"max_gift\", \"mean_gift\" and \"min_gift\"\n", 431 | "next_variable = next_best([\"max_gift\", \"mean_gift\", \"min_gift\"], [\"age\", \"gender_F\"], [\"target\"], basetable)\n", 432 | "print(next_variable)\n", 433 | "\n", 434 | "# Calculate the AUC of a model that uses \"max_gift\", \"mean_gift\", \"min_gift\" and \"age\" as predictors\n", 435 | "auc_current_age = auc([\"max_gift\", \"mean_gift\", \"min_gift\", \"age\"], [\"target\"], basetable)\n", 436 | "print(round(auc_current_age,4))\n", 437 | "\n", 438 | "# Calculate the AUC of a model that uses \"max_gift\", \"mean_gift\", \"min_gift\" and \"gender_F\" as predictors\n", 439 | "auc_current_gender_F = auc([\"max_gift\", \"mean_gift\", \"min_gift\", \"gender_F\"], [\"target\"], basetable)\n", 440 | "print(round(auc_current_gender_F,4))" 441 | ] 442 | }, 443 | { 444 | "cell_type": "markdown", 445 | "metadata": {}, 446 | "source": [ 447 | "#### The model that has `gender_F` as next variable has a better AUC than the model that has `age` as next variable. Therefore, `gender_F` is selected as the next best variable." 448 | ] 449 | }, 450 | { 451 | "cell_type": "markdown", 452 | "metadata": {}, 453 | "source": [ 454 | "## Finding the order of variables\n", 455 | "- The forward stepwise variable selection procedure starts with an empty set of variables, and adds predictors one by one. In each step, the predictor that has the highest AUC in combination with the current variables is selected. \n", 456 | "- **Task** : implement the forward stepwise variable selection procedure. To this end, we can use the next_best function " 457 | ] 458 | }, 459 | { 460 | "cell_type": "code", 461 | "execution_count": 26, 462 | "metadata": {}, 463 | "outputs": [ 464 | { 465 | "name": "stdout", 466 | "output_type": "stream", 467 | "text": [ 468 | "['gender_F', 'income_high', 'income_low', 'country_USA', 'country_India', 'country_UK', 'age', 'time_since_last_gift', 'time_since_first_gift', 'max_gift', 'min_gift', 'mean_gift', 'number_gift']\n" 469 | ] 470 | } 471 | ], 472 | "source": [ 473 | "# Find the candidate variables\n", 474 | "candidate_variables = list(basetable.columns.values)\n", 475 | "candidate_variables.remove(\"target\")\n", 476 | "print(candidate_variables)" 477 | ] 478 | }, 479 | { 480 | "cell_type": "code", 481 | "execution_count": 28, 482 | "metadata": {}, 483 | "outputs": [ 484 | { 485 | "name": "stdout", 486 | "output_type": "stream", 487 | "text": [ 488 | "Variable added in step 1 is time_since_last_gift.\n", 489 | "Variable added in step 2 is age.\n", 490 | "Variable added in step 3 is country_UK.\n", 491 | "Variable added in step 4 is country_India.\n", 492 | "Variable added in step 5 is country_USA.\n", 493 | "['time_since_last_gift', 'age', 'country_UK', 'country_India', 'country_USA']\n" 494 | ] 495 | } 496 | ], 497 | "source": [ 498 | "# Initialize the current variables\n", 499 | "current_variables = []\n", 500 | "\n", 501 | "# The forward stepwise variable selection procedure\n", 502 | "number_iterations = 5\n", 503 | "for i in range(0, number_iterations):\n", 504 | " next_variable = next_best(current_variables, candidate_variables, [\"target\"], basetable)\n", 505 | " current_variables = current_variables + [next_variable]\n", 506 | " candidate_variables.remove(next_variable)\n", 507 | " print(\"Variable added in step \" + str(i+1) + \" is \" + next_variable + \".\")\n", 508 | "print(current_variables)" 509 | ] 510 | }, 511 | { 512 | "cell_type": "markdown", 513 | "metadata": {}, 514 | "source": [ 515 | "## Correlated variables\n", 516 | "- The first 10 variables that are added to the model are the following: \n", 517 | "`['max_gift', 'number_gift', 'time_since_last_gift', 'mean_gift', 'income_high', 'age', 'country_USA', 'gender_F', 'income_low', 'country_UK']`\n", 518 | "- `min_gift` is not added. Does this mean that it is a bad variable? We can test the performance of the variable by using it in a model as a single variable and calculating the AUC. How does the AUC of `min_gift` compare to the AUC of `income_high`? To this end, we can use the function auc()\n", 519 | "- It can happen that a **good variable is not added because it is highly correlated with a variable that is already in the model**. We can test this calculating the correlation between these variables\n", 520 | "\n", 521 | "```python\n", 522 | "import numpy\n", 523 | "numpy.corrcoef(basetable[\"variable_1\"],basetable[\"variable_2\"])[0,1]\n", 524 | "```" 525 | ] 526 | }, 527 | { 528 | "cell_type": "code", 529 | "execution_count": 29, 530 | "metadata": {}, 531 | "outputs": [], 532 | "source": [ 533 | "import numpy as np" 534 | ] 535 | }, 536 | { 537 | "cell_type": "markdown", 538 | "metadata": {}, 539 | "source": [ 540 | "- Calculate the AUC of the model using the variable `min_gift` only.\n", 541 | "- Calculate the AUC of the model using the variable `income_high` only.\n", 542 | "- Calculate the correlation between the variable `min_gift` and `mean_gift`." 543 | ] 544 | }, 545 | { 546 | "cell_type": "code", 547 | "execution_count": 30, 548 | "metadata": {}, 549 | "outputs": [ 550 | { 551 | "name": "stdout", 552 | "output_type": "stream", 553 | "text": [ 554 | "0.57\n", 555 | "0.52\n", 556 | "0.76\n" 557 | ] 558 | } 559 | ], 560 | "source": [ 561 | "# Calculate the AUC of the model using min_gift only\n", 562 | "auc_min_gift = auc([\"min_gift\"], [\"target\"], basetable)\n", 563 | "print(round(auc_min_gift,2))\n", 564 | "\n", 565 | "# Calculate the AUC of the model using income_high only\n", 566 | "auc_income_high = auc([\"income_high\"], [\"target\"], basetable)\n", 567 | "print(round(auc_income_high,2))\n", 568 | "\n", 569 | "# Calculate the correlation between min_gift and mean_gift\n", 570 | "correlation = np.corrcoef(basetable[\"min_gift\"], basetable[\"mean_gift\"])[0,1]\n", 571 | "print(round(correlation,2))" 572 | ] 573 | }, 574 | { 575 | "cell_type": "markdown", 576 | "metadata": {}, 577 | "source": [ 578 | "#### `min_gift` has more predictive power than `income_high`, but that it is highly correlated with `mean_gift` and therefore not included in the selected variables." 579 | ] 580 | }, 581 | { 582 | "cell_type": "markdown", 583 | "metadata": {}, 584 | "source": [ 585 | "### Partitioning\n", 586 | "- In order to properly evaluate a model, one can partition the data in a train and test set. The train set contains the data the model is built on, and the test data is used to evaluate the model. This division is done randomly, but when the target incidence is low, it could be necessary to stratify, that is, to make sure that the train and test data contain an equal percentage of targets. \n", 587 | "- **Task** : partition the data with stratification and verify that the train and test data have equal target incidence." 588 | ] 589 | }, 590 | { 591 | "cell_type": "code", 592 | "execution_count": 35, 593 | "metadata": {}, 594 | "outputs": [], 595 | "source": [ 596 | "# Load the partitioning module\n", 597 | "from sklearn.model_selection import train_test_split" 598 | ] 599 | }, 600 | { 601 | "cell_type": "code", 602 | "execution_count": 40, 603 | "metadata": {}, 604 | "outputs": [ 605 | { 606 | "name": "stdout", 607 | "output_type": "stream", 608 | "text": [ 609 | "0.05\n", 610 | "0.05\n" 611 | ] 612 | } 613 | ], 614 | "source": [ 615 | "# Create dataframes with variables and target\n", 616 | "X = basetable.drop(\"target\", 1)\n", 617 | "y = basetable[\"target\"]\n", 618 | "\n", 619 | "# Carry out 50-50 partititioning with stratification\n", 620 | "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.5, stratify = y)\n", 621 | "\n", 622 | "# Create the final train and test basetables\n", 623 | "train = pd.concat([X_train, y_train], axis=1)\n", 624 | "test = pd.concat([X_test, y_test], axis=1)\n", 625 | "\n", 626 | "# Check whether train and test have same percentage targets\n", 627 | "print(round(sum(train['target'])/len(train), 2))\n", 628 | "print(round(sum(test['target'])/len(test), 2))" 629 | ] 630 | }, 631 | { 632 | "cell_type": "markdown", 633 | "metadata": {}, 634 | "source": [ 635 | "#### The stratify option makes sure the target incidence is the same in both train and test." 636 | ] 637 | }, 638 | { 639 | "cell_type": "markdown", 640 | "metadata": {}, 641 | "source": [ 642 | "### Evaluating a model on test and train\n", 643 | "- **Task** : apply AUC function, and check whether the train and test AUC are similar.\n", 644 | "- Calculate the train and test AUC of the model using `\"age\"` and `\"gender_F\"` as predictors using the auc_train_test function." 645 | ] 646 | }, 647 | { 648 | "cell_type": "code", 649 | "execution_count": 43, 650 | "metadata": {}, 651 | "outputs": [ 652 | { 653 | "name": "stdout", 654 | "output_type": "stream", 655 | "text": [ 656 | "0.55\n", 657 | "0.53\n" 658 | ] 659 | } 660 | ], 661 | "source": [ 662 | "# Carry out 70-30 partititioning with stratification\n", 663 | "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.3, stratify = y)\n", 664 | "\n", 665 | "# Create the final train and test basetables\n", 666 | "train = pd.concat([X_train, y_train], axis=1)\n", 667 | "test = pd.concat([X_test, y_test], axis=1)\n", 668 | "\n", 669 | "auc_train = auc([\"age\", \"gender_F\"], \"target\", train)\n", 670 | "auc_test = auc([\"age\", \"gender_F\"], \"target\", test)\n", 671 | "\n", 672 | "print(round(auc_train,2))\n", 673 | "print(round(auc_test,2))" 674 | ] 675 | }, 676 | { 677 | "cell_type": "markdown", 678 | "metadata": {}, 679 | "source": [ 680 | "#### It could happen that the test AUC is slightly lower than the train AUC. This is a perfectly normal phenomenon called over-fitting." 681 | ] 682 | }, 683 | { 684 | "cell_type": "markdown", 685 | "metadata": {}, 686 | "source": [ 687 | "### Building the AUC curves\n", 688 | "- **The forward stepwise variable selection procedure provides an order in which variables are optimally added to the predictor set. In order to decide where to cut off the variables, we can make the train and test AUC curves. These curves plot the train and test AUC using the first, first two, first three, ... variables in the model.**\n", 689 | "- **Task** : plot these AUC curves\n", 690 | "- variables = ['max_gift',\n", 691 | " 'time_since_last_gift',\n", 692 | " 'number_gift',\n", 693 | " 'mean_gift',\n", 694 | " 'income_high',\n", 695 | " 'age',\n", 696 | " 'gender_F',\n", 697 | " 'time_since_first_gift',\n", 698 | " 'income_low',\n", 699 | " 'country_UK']\n", 700 | " \n", 701 | "- `auc_values_train`, will contain the train AUC values of the model at each iteration\n", 702 | "- `auc_values_test`, will contain the test AUC values of the model at each iteration\n", 703 | "- `variables_evaluate`, will contain the variables evaluated at each iteration\n", 704 | "\n", 705 | "**************************************************************\n", 706 | "- Iterate over the variables.\n", 707 | "- In each iteration, add the next variable in variables to variables_evaluate.\n", 708 | "- In each iteration, calculate the train and test AUC using the auc method. The dataframes train and test contain the train and test data respectively.\n", 709 | "- In each iteration, add the calculated values to auc_values_train and auc_values_test" 710 | ] 711 | }, 712 | { 713 | "cell_type": "code", 714 | "execution_count": 47, 715 | "metadata": {}, 716 | "outputs": [ 717 | { 718 | "data": { 719 | "image/png": "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\n", 720 | "text/plain": [ 721 | "
" 722 | ] 723 | }, 724 | "metadata": { 725 | "needs_background": "light" 726 | }, 727 | "output_type": "display_data" 728 | } 729 | ], 730 | "source": [ 731 | "# Keep track of train and test AUC values\n", 732 | "auc_values_train = []\n", 733 | "auc_values_test = []\n", 734 | "variables_evaluate = []\n", 735 | "\n", 736 | "variables = ['max_gift',\n", 737 | " 'time_since_last_gift',\n", 738 | " 'number_gift',\n", 739 | " 'mean_gift',\n", 740 | " 'income_high',\n", 741 | " 'age',\n", 742 | " 'gender_F',\n", 743 | " 'time_since_first_gift',\n", 744 | " 'income_low',\n", 745 | " 'country_UK']\n", 746 | "\n", 747 | "# Iterate over the variables in variables\n", 748 | "for v in variables:\n", 749 | " \n", 750 | " # Add the variable\n", 751 | " variables_evaluate.append(v)\n", 752 | " \n", 753 | " # Calculate the train and test AUC of this set of variables\n", 754 | " auc_train = auc(variables_evaluate, \"target\", train)\n", 755 | " auc_test = auc(variables_evaluate, \"target\", test)\n", 756 | " \n", 757 | " # Append the values to the lists\n", 758 | " auc_values_train.append(auc_train)\n", 759 | " auc_values_test.append(auc_test)\n", 760 | " \n", 761 | "# Make plot of the AUC values\n", 762 | "import matplotlib.pyplot as plt\n", 763 | "import numpy as np\n", 764 | "\n", 765 | "x = np.array(range(0,len(auc_values_train)))\n", 766 | "y_train = np.array(auc_values_train)\n", 767 | "y_test = np.array(auc_values_test)\n", 768 | "plt.xticks(x, variables, rotation = 90)\n", 769 | "plt.plot(x,y_train)\n", 770 | "plt.plot(x,y_test)\n", 771 | "plt.ylim((0.6, 0.8))\n", 772 | "plt.show()" 773 | ] 774 | }, 775 | { 776 | "cell_type": "markdown", 777 | "metadata": {}, 778 | "source": [ 779 | "#### Note that the test AUC curve starts declining sooner than the train curve. The point at which this happens is a good cut-off." 780 | ] 781 | } 782 | ], 783 | "metadata": { 784 | "kernelspec": { 785 | "display_name": "Python 3", 786 | "language": "python", 787 | "name": "python3" 788 | }, 789 | "language_info": { 790 | "codemirror_mode": { 791 | "name": "ipython", 792 | "version": 3 793 | }, 794 | "file_extension": ".py", 795 | "mimetype": "text/x-python", 796 | "name": "python", 797 | "nbconvert_exporter": "python", 798 | "pygments_lexer": "ipython3", 799 | "version": "3.7.3" 800 | } 801 | }, 802 | "nbformat": 4, 803 | "nbformat_minor": 2 804 | } 805 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Predictive-Analytics-in-Python 2 | - Build ML model with meaningful variables. Use model for predictions. 3 | - **Predictive analytics** is an process that aims at predicting an event using historical data. This data is gathered in the analytical basetable. 4 | 5 | ### Analytical Basetable structure 6 | - An **analytical base table** is typically stored in a pandas dataframe. Three important variables in the analytical basetable are : **`population`**, **`candidate predictors`** and the **`target`** 7 | - **Population** is the group of people or object we want to make the predicton for **(rows of data)** 8 | - **Candidate predictors** is the information that can be used to predict the event **(features)** 9 | - **Target** is the event to **`predict`**, 10 | 11 | ```python 12 | import pandas as pd 13 | basetable = pd.DataFrame("import_basetable.csv") 14 | population_size = len(basetable) 15 | targets = sum(basetable["target"]) 16 | ``` 17 | 18 | ### Logistic Regression 19 | 20 | ```python 21 | from sklearn import linear_model 22 | logreg = linear_model.LogisticRegression() 23 | X = basetable[['age']] 24 | y = basetable[['target']] 25 | logreg.fit(X, y) 26 | print(logreg.coef_) 27 | print(logreg.intercept_) 28 | 29 | ``` 30 | 31 | ### Multivariate logistic regression 32 | - Univariate : ax+b 33 | - Multivariate : a1x1 + a2x2 + a3x3 +....+ anxn + b 34 | 35 | ### Making predictions 36 | 37 | ```python 38 | new_data = current_data[["gender_F", "ag", "time_since_last_gift"]] 39 | predictions = logreg.predict(new_data) 40 | ``` 41 | 42 | ### Variable selection 43 | 44 | #### Drawbacks of models with many variables 45 | - Overfitting 46 | - Hard to maintain or implement 47 | - Hard to interpret, multi-collinearity : correalted variables make interpretation harder 48 | 49 | - The goal of variable selection is to select a set of variables that has optimal performance 50 | 51 | ### Model evaluation : AUC 52 | - A metric often used to quantify the model performance is AUC value. It is a value between 0 & 1, "1" being the perfect model. 53 | 54 | ```python 55 | from sklearn.metrics import roc_auc_score 56 | roc_auc_score(true_target, prob_target) 57 | ``` 58 | 59 | - **`the model with 5 variables has the same AUC as the model using only 2 variables. Adding more variables doesn't always increase the AUC`** 60 | 61 | - **`AUC score can be used to determine whether increasing or decreasing the model variables increases the performance or not`** 62 | 63 | 64 | ### Forward stepwise variable selection : Intutive way of variable selection 65 | - The forward stepwise variable selection procedure: 66 | **First** it selects among all the candidate predictors the variable that has the best AUC when used in the model. **Next**, it selects another candidate predictor that has the best AUC incombination with the first selected variable. This continues until all variables are added or until predefined number of variables is added. 67 | - Find best variable **v1** 68 | - Find best variable **v2** in combination with **v1** 69 | - Find best variable **v3** in combination with **v1,v2** 70 | - **Until all variables are added or until predefined number of variables is added** 71 | 72 | ### Implementation of the forward stepwise variable selection 73 | - Function **auc** that calculates AUC given a certain set of variables 74 | - Function **best_next** that returns next best variable in combination with current variables 75 | - Loop until desired number of variables 76 | 77 | 78 | ### Implementation of AUC function 79 | 80 | ```python 81 | from sklearn import linear_model 82 | from sklearn.metrics import roc_auc_score 83 | 84 | def auc(variables, target, basetable): 85 | X = basetable[variables] 86 | y = basetable[target] 87 | 88 | logreg = linear_model.LogisticRegression() 89 | logreg.fit(X,y) 90 | 91 | predictions = logreg.predict_proba(X)[:,1] 92 | auc = roc_auc_score(y, predictions) 93 | return auc 94 | ``` 95 | 96 | - calling **auc** 97 | 98 | ```python 99 | auc = auc(["age","gender_F"], ["target"], basetable) 100 | print(round(auc, 2)) 101 | ``` 102 | 103 | ### Calculating the next best vriable 104 | 105 | ```python 106 | def next_best(current_variable, candidate_variable, target, basetable): 107 | """function looks throughout candidate variables and keeps track of which variable is best and the auc associated with the best variable""" 108 | best_auc = -1 109 | best_variable = None 110 | 111 | # for each variable in the candidate variable set calculate the AUC 112 | # current_variable : variables which are already in the model 113 | # extend it with the variable with which we need to evaluate 114 | for v in candidate_variables: 115 | auc_v = auc(current_variables + [v], target, basetable) 116 | 117 | # if this AUC is better then the best AUC, change the best AUC and best variable 118 | if auc_v >= best_auc: 119 | best_auc = auc_v 120 | best_variable = v 121 | return best_variable 122 | ``` 123 | 124 | - If we want to know which variable among `min_gift, max_gift, mean_gift` should be added next given that `age and gender_F` are already in the model, we can use **`next_best`** function as follows; 125 | 126 | ```python 127 | current_variables = ["age", "gender_F"] 128 | candidate_variables = ["min_gift", "max_gift", "mean_gift"] 129 | next_variable = next_best(current_variables, candidate_variables, basetable) 130 | print(next_varible) 131 | ``` 132 | 133 | - To complete the forward stepwise variable selection procedure, we keep track of the candidate variables and current variables added to the model so far 134 | 135 | ```python 136 | candidate_variables = ["mean_gift", "min_gift", "max_gift","age","gender_F", "country_USA", "income_low"] 137 | current_variables = [] 138 | target = ["target"] 139 | ``` 140 | 141 | - We can define the max number of variables that can be added. In each iteration, the next_best variable is calculated using the next_best function. The current variable list is updated by already chosen variable and the chosen variable is removed from the candidate variable list. 142 | 143 | ```python 144 | max_number_variables = 5 145 | number_iterations = min(max_number_variables, len(candidate_variables)) 146 | for i in range(0, number_iterations): 147 | next_var = next_best(current_variables, candidate_variables, target, basetable) 148 | current_variables = current_variables + [next_variable] 149 | candidate_variables.remove(next_variable) 150 | 151 | print(current_variables) 152 | ``` 153 | 154 | ### Deciding on the number of variables 155 | - Forward Stepwise variable selection returns the order in which the variables increase the accuracy, but we need to decide on how many variables to use. 156 | 157 | ```python 158 | auc_values = [] 159 | variables_evaluate = [] 160 | 161 | for v in variables_forward: 162 | variables_evaluate.append(v) 163 | auc_value = auc(variables_evaluate, ["target"], basetable) 164 | auc_values.append(auc_value) 165 | ``` 166 | 167 | - Inorder to do so, we can have a look at the **AUC values**. The order of the variables is given in the list `variables_forward`. For each variable in variable forward calculate the AUC values. 168 | 169 |

170 | 171 |

172 | 173 | - If we plot the AUC values we obtain a curve that typically keeps increasing 174 | . However, if we use new data to evaluate subsequent models it doesn't increase, instead it decreases after a while. This phenomenon is called overfitting. 175 | - By adding more variables the accuracy on the data on which model is built increases, but the true performance of the model decreases because the complex model doesnt generalize to other data. 176 | 177 | #### Detecting over-fitting 178 | - There exits smart techniques to detect and prevent overfitting. Performance on the test dataset is representative of the true performance of the model. 179 | - One way of partioning data is randomly dividing the data into two parts, however when the data is imbalanced it is important to make sure that the target variable is in same proportion in train and test. It can be done by using **`stratify`** on the target while splitting the data. 180 | 181 | ```python 182 | from sklearn.model_selection import train_test_split 183 | 184 | X = basetable.drop("target", 1) 185 | y = basetable["target"] 186 | 187 | X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, stratify = Y) 188 | 189 | train = pd.concat([X_train, y_train], axis=1) 190 | test = pd.concat([X_test, y_test], axis=1) 191 | ``` 192 | 193 | #### Deciding the cut-off 194 | 195 |

196 | 197 |

198 | 199 | - We can now plot AUC curves of the subsequent models on both the train and test data.We can see that the train AUC keeps increasing while the test AUC stabalizes and then decreases. 200 | - When deciding on how many variables to keep in the model, once we take into account that **test AUC is as high as possible** and the **model should have least variables possible**. 201 | - In this case it's clear that the cut-off indicated by the dashed line is the best option. All models having more variables has lower test accuracy. 202 | 203 | ## Explaining model performance to business 204 | 205 | ### The cumulative gains curve 206 | - Once the model is ready we need to show it to the business. VIsualization of model performance that business people can understand. 207 | - Until now we evaluated models using the AUC. Though it is very useful for data scientist it is less appropriate if we want to discuss the model performance with business stakeholders. 208 | - Indeed, AUC is a bit complex evaluation measure that is not much intutive.Moreover, its a single number which doesn't catch all the information about the model. 209 | - For better visualization we can use evaluation curve like the cumulative gains curve. This type of curves are easy to explain and guide us to better business decisions. 210 | 211 | #### **Cumulative gains curve** is constructed as follows : 212 | 213 |

214 | 215 |

216 | 217 | - First, we order all the observations according to the output of the model. One the LHS are the observations with the highest probabilty to be target according to the model and on the RHS are the observations with lowest probabilty to be target. 218 | - On the horizontal axis of cumulative gains curve, it is indicated which percentage of the observations is considered. For instance, 30% of the observations with the highest probabilty to be target is considered. 219 | - On the vertical axis, the curve indicates which percentage of all targets is included in this group. For instance, if the cumulative gain is 70% at 30%, it means we are taking the top 30% observations with highest probabilty to be target, this group contains already 70% of all targets. 220 | 221 |

222 | 223 |

224 | 225 | - The cumulative gains curve is the great tool to compare models. **The more the line is situated to the upper left corner, the better the model**. It is often the case that two models produce curves that cross each other. In that case, it is not straightforward to decide which model is best. In this case, for instance, we can say that model 2 is better to distinguish the top 10% observations from the rest, while model 1 is better to distinguish the top 70% of the observations from the rest. 226 | 227 | #### Cumulative gains in python 228 | - Constructing cumulative gains curves in Python is easy with the **scikitplot** module. 229 | 230 | ```python 231 | import scikitplot as skplt 232 | import matplotlib.pyplot as plt 233 | 234 | skplt.metrics.plot_cumulative_gain(true_values, predictions) 235 | plt.show() 236 | ``` 237 | - **Note that predictions should have values for both targets 1 and 0.** We can use `predict_proba` to get prediction probablities for both the targets.README 238 | 239 | ### The Lift Curve 240 | - In addition to the cumulative gains curve, the lift curve is a widely used visualisation of model performance. 241 | 242 |

243 | 244 |

245 | 246 | - Constructing a lift curve follows a similar process as constructing a cumulative gains curve. First, we order all the observations according to the model. On the horizontal axis of the lift curve, we indicate which percentage of the observation is considered. On the vertical axis, the lift curve indicates how many times more than average targets are included in this group. 247 | - Consider for instance the lift at 10%, and assume that the top 10% observations contains 20% targets. If the average percentage of targets is 5%, the lift is 4, because 20% is 4 times 5%. 248 | - As another example, consider the lift at 50%, and assume that the top 50% observations contain 10% targets. As 10% is 2 times 5%, the average percentage of targets, the lift is 2 at 50%. 249 | 250 | #### Lift curve interpretation 251 | 252 |

253 | 254 |

255 | 256 | - A random model has a more or less equal percentage of targets for each group, and therefore the baseline is 1. Better models has higher lifts. Therefore **curves that are higher, have better accuracy**. However, as for cumulative gains curve, also lift curves of different models can cross each other. 257 | - Consider the example given here: model 2 is higher at 10%, but model 1 is higher at 80%. In that case it is hard to say which model is best, as it depends on the situation.If we can target 10% of the population, model 2 is better suited because we can reach more targets, whereas model 1 is better if we want to target a larger part of the population. 258 | 259 | #### Lift curve in Python 260 | 261 | ```python 262 | import scikitplot as skplt 263 | import matplotlib.pyplot as plt 264 | 265 | skplt.metrics.plot_lift_curve(true_values, predictions) 266 | plt.show() 267 | ``` 268 | 269 | ### Guiding business to better decisions 270 | - One way to make use of lift graph is to estimate the profit for a campaign 271 | 272 | #### Estimating profit 273 | - The population consists of 100000 candidate donors, and 5% among these candidate donors is target. Assume that we expect targets to donate 50 Euro, and that addressing a donor, for instance by sending him a letter, costs 2 Euro. Given this info we can calculate the expected profit of a campaign. 274 | - The profit depends on 5 elements: the percentage of targets in the selected donor group, the percentage selected donors, the population size and the reward of reaching a target and the cost of the campaign. 275 | - The total cost of the campaign is the cost of the campaign, 2 Euro, times the no of donors addressed, which is the percentage of selected donors times the population size. 276 | - The reward of the campaign is the reward of reaching a target times the number of targets reached. The final profit is then the reward minus the cost. 277 | - Assume that we address the top 20% donors with the highest probabilty to donate according to the model. 278 | 279 | 280 | ```python 281 | population = 100000 282 | target_incidence = 0.05 283 | reward_target = 50 284 | cost_campaign = 2 285 | 286 | def profit(perc_targets, perc_selected, population_size, reward_target, cost_campaign) 287 | cost = cost_campaign * perc_selected * population_size 288 | reward = reward_target * perc_targets * perc_selected * population_size 289 | return (reward - cost) 290 | 291 | perc_selected = 0.20 292 | lift = 2.5 293 | perc_targets = lift * target_incidence 294 | print(profit(perc_targets, perc_selected, population_size, reward_target, cost_campaign)) 295 | ``` 296 | 297 | 298 | ## Interpreting and explaining models 299 | 300 | ### Predictor Insights Graphs 301 | - It is important to check with business and domain experts whether the model is interpretable. In a typical predictive modelling project, we proceed as follows when we need to make a prediction model: 302 | - **1)** First, we contruct a predictive model (Build model) 303 | - **2)** Then, we can evaluate the predictive model using the AUC accuracy metric, and additionally using the cumulative gains and lift curves. 304 | - **3)** One last step that we should carry out to make sure that the model is sound and logical, is to interpret the variables in our model, and verify whether the link between these variables and the target makes sense. (Verify whether the variables in the model are interpretable) 305 | 306 | - Here we can use the **predictor insights graphs** 307 | 308 | #### Predictor insights graphs for categorical variable 309 | 310 |

311 | 312 |

313 | 314 | - **These graphs shows the link between the predictive variables and the target that we want to predict**. 315 | - Consider for instance this predictor insight graph that shows the relationship between income and donations. On the horizontal axis, the predictive variable is divided into 3 groups, donors with low, average and high income. The height of the grey bars indicate how many donors are in each group and is associated with the left hand side vertical axis. The green line indicates the perentage of targets in each group and is associated with the RHS vertical axis. 316 | - In this graph, we can see that the higher someones income is the more likely he is to donate for the campaign. **This interpretation is logical, so it makes sense to keep variables related to income in the model.** 317 | 318 | #### Predictor insights graphs for continous variable 319 | 320 |

321 | 322 |

323 | 324 | - If the variable is continuos, an additional **discretization step that divides the continuos variables in bins** is needed. Above example shows the relationship between the time since someone first donated and the target. 325 | - The continuous variable `days since first donation` is split in five groups of equal size, and then the size of each group and target incidence is plot for each group. **It shows that the longer someone is a donor, the more likely he is to donate for the campaign.** 326 | 327 | #### The predictor insight graph table 328 | 329 | |Income |size | Incidence| 330 | :--------:|:------:|:--------:| 331 | |low |20850 |0.0431 | 332 | |average |62950 |0.0492 | 333 | |high |16200 |0.0615 | 334 | 335 | - The values that are plotted in the predictor insight graph, is collected in a predictor insight graph table. This table has 3 columns: the categories that are displayed on the horizontal axis, the size of the group as displayed on LHS axis and the target incidence of each group displayed on the RHS axis. 336 | - We can access elements in the predictor insight graph using indexing 337 | 338 | ```python 339 | print(pig_table["size"][income=="low"]) 340 | ``` 341 | 342 | ### Constructing a predictor insights graph 343 | - **1)** If the variable at hand is continous, first discretize the variable 344 | - **2)** Next, calculate the values that are needed to make the plot, these values are gathered in the PIG table (Calculate predictor insight graph table) 345 | - **3)** Finally, plot the predictor insight graph 346 | 347 | ### Preparing the predictor insights graph table 348 | 349 | ```python 350 | import numpy as np 351 | 352 | # function that calculates the predictor insight graph table 353 | def create_pig_table(df, target, variable): 354 | 355 | # group by the variable we want to plot 356 | groups = df[[target, variable]].groupby(variable) 357 | 358 | #calculate the size and incidence of each group 359 | pig_table = groups[target].agg({'Incidence':np.mean, 'size': np.size}).reset_index() 360 | return pig_table 361 | print(create_pig_table(basetable, "target", "country") 362 | ``` 363 | 364 | - Construct the table that is needed to plot the predictor insight graphs.A PIG table is a table that has all the information necessary to create the predictor insight graph.It has one row for each group in the variable that we want to plot, and three columns.The first column contains the names of the groups, incase the original variable is continuous, these are the names of intervals it was discretized in. The second column shows the average target incidence of the group:what is the mean target in this group.The 3rd column shows the size of each group i.e the number of observations that belongs to the particular group. 365 | - We can easily construct a predictor insight graph for a given basetable, target and variable.In `create_pig_table` function we first group the basetable by the variable we want to make the predictor insight graph for. In these groups, we only need the variable and target values, that is why we only select these in this step.Next we use the aggregate function on these groups to create two columns.The first column is the target incidence, which is the mean of the target,2nd column is the size i.e the no of observations in each group. With this function we can easily calculate the predictor insight graph table for any variable. 366 | 367 | #### Calculating multiple predictor insight graph tables 368 | - Instead of calculating them one by one, we could do this automatically and store the PIG tables in a dictionary. 369 | 370 | ```python 371 | # variable to plot 372 | variables = ['country', 'gender', 'disc_mean_gift', 'age'] 373 | 374 | # empty dictionary 375 | pig_tables = {} 376 | 377 | #loop over all variables 378 | for variable in variables: 379 | 380 | # create the PIG table 381 | pig_table = create_pig_table(basetable, "target", variable) 382 | 383 | # store the table in the dict 384 | pig_table[variable] = pig_table 385 | ``` 386 | 387 | - We can print PIG for any variable using the `pig_tables` dictionary. 388 | 389 | ### Plotting the predictor insight graph 390 | - The PIG shows the link between a predictor and the target. For instance, the predictor insight graph of the predictor gender shows that females are more likely to donate.Additionally, this graph also shows the size of the different groups. 391 | - We construct the graph in two steps: First, we plot the target incidence line, and secondly we will plot the bars that show the group sizes as well. The values needed to contruct the PIG can be obtained from the PIG table. 392 | - Inorder to add the graphs with the sizes we need to add few lines of code 393 | 394 | ```python 395 | import matplotlib.pyplot as plt 396 | import numpy as np 397 | 398 | # plot the graph 399 | plt.ylabel("size", rotation=0, rotation_mode="anchor", ha="right") 400 | 401 | pig_table["Incidence"].plot(secondary_y = True) 402 | 403 | pig_table["Size"].plot(kind='bar', width=0.5, color="lightgray", edgecolor="none") 404 | 405 | #show the group names 406 | plt.xticks(np.arange(len(pig_table)), pig_table['income']) 407 | 408 | #center the group names by adding margin to LHS and RHS of the plot 409 | width=0.5 410 | plt.xlim([-width, len(pig_table) - width]) 411 | 412 | #add label incidence to the vertical axis 413 | plt.ylabel("incidence", rotation = 0, rotation_mode="anchor", ha="right") 414 | 415 | # add label income to the horizontal axis 416 | plt.xlabel("Income") 417 | plt.show() 418 | ``` 419 | 420 | ### Selecting relevant data (Timeline of data) 421 | 422 | ```python 423 | import pandas as pd 424 | 425 | data['date'] = pd.to_datetime(data['date']) 426 | 427 | start_target = datetime(year = 2018, month = 5, day = 1) 428 | end_target = datetime(year = 2018, month = 8, day = 1) 429 | 430 | gifts_target = gifts[(gifts['date'] >= start_target) & (gifts['date'] < end_target)] 431 | ``` 432 | 433 | gifts_pred_variables = gifts[(gifts['date'] < start_target)] 434 | ``` 435 | 436 | 437 | 438 | 439 | 440 | 441 | 442 | 443 | 444 | 445 | 446 | 447 | 448 | 449 | 450 | 451 | 452 | 453 | 454 | 455 | 456 | 457 | 458 | 459 | 460 | 461 | 462 | 463 | 464 | 465 | 466 | 467 | 468 | 469 | 470 | 471 | 472 | 473 | 474 | 475 | 476 | 477 | 478 | 479 | 480 | 481 | 482 | 483 | 484 | 485 | 486 | 487 | 488 | 489 | 490 | 491 | 492 | 493 | 494 | 495 | 496 | 497 | 498 | 499 | 500 | 501 | 502 | 503 | 504 | 505 | 506 | 507 | 508 | 509 | 510 | 511 | 512 | 513 | 514 | 515 | 516 | 517 | 518 | 519 | 520 | 521 | 522 | 523 | 524 | 525 | 526 | 527 | 528 | 529 | 530 | 531 | 532 | 533 | 534 | 535 | 536 | 537 | 538 | 539 | 540 | 541 | 542 | 543 | 544 | 545 | 546 | 547 | 548 | 549 | 550 | 551 | 552 | 553 | 554 | 555 | 556 | 557 | 558 | 559 | 560 | 561 | 562 | 563 | 564 | 565 | 566 | 567 | 568 | 569 | 570 | 571 | 572 | 573 | 574 | 575 | 576 | 577 | 578 | 579 | 580 | 581 | 582 | 583 | 584 | 585 | 586 | 587 | 588 | 589 | 590 | 591 | 592 | 593 | 594 | 595 | 596 | 597 | 598 | 599 | 600 | 601 | 602 | 603 | 604 | 605 | 606 | 607 | 608 | 609 | 610 | 611 | 612 | 613 | 614 | 615 | 616 | 617 | 618 | 619 | 620 | 621 | 622 | 623 | 624 | 625 | 626 | 627 | 628 | 629 | 630 | 631 | 632 | 633 | 634 | 635 | 636 | 637 | 638 | 639 | 640 | 641 | 642 | 643 | 644 | 645 | 646 | 647 | 648 | 649 | 650 | 651 | 652 | 653 | 654 | 655 | 656 | 657 | 658 | 659 | 660 | 661 | 662 | 663 | 664 | 665 | 666 | 667 | 668 | 669 | 670 | 671 | 672 | 673 | 674 | 675 | 676 | 677 | 678 | 679 | 680 | 681 | 682 | 683 | 684 | 685 | 686 | 687 | 688 | 689 | 690 | 691 | 692 | 693 | 694 | 695 | 696 | 697 | 698 | 699 | 700 | 701 | 702 | 703 | 704 | 705 | 706 | 707 | 708 | 709 | 710 | 711 | 712 | 713 | 714 | 715 | 716 | 717 | 718 | 719 | 720 | 721 | 722 | 723 | 724 | 725 | 726 | -------------------------------------------------------------------------------- /classification/spill_incidents_analytics/README.md: -------------------------------------------------------------------------------- 1 | ### Spill Incidents Analytics 2 | 3 | - `emerson_spill_incidents_eda.ipynb` : EDA of spill dataset 4 | - `emerson_spill_incidents_modelling.ipynb` : contains model building 5 | 6 | #### Instructions 7 | 8 | - Create new python virtual environment `python -m venv envname` 9 | - `pip install -r requirements.txt` 10 | 11 | #### Note : Tried to build a first cut solution. We can do lot of more fine tunings/enhancements. -------------------------------------------------------------------------------- /classification/spill_incidents_analytics/requirements.txt: -------------------------------------------------------------------------------- 1 | 2 | ##### Core scientific packages 3 | jupyter==1.0.0 4 | matplotlib==3.3.4 5 | numpy==1.19.5 6 | pandas==1.2.2 7 | scipy==1.6.0 8 | plotly==4.14.3 9 | statsmodels==0.12.2 10 | wordcloud==1.8.1 11 | shap==0.39.0 12 | 13 | ##### Machine Learning packages 14 | scikit-learn==0.24.1 15 | xgboost==1.3.3 16 | 17 | ##### TensorFlow-related packages 18 | 19 | tensorflow==1.15.5 # or tensorflow-gpu==1.15.5 for GPU support 20 | 21 | tensorboard==1.15.0 22 | 23 | # There are a few dependencies you need to install first, check out: 24 | # https://github.com/openai/gym#installing-everything 25 | gym[atari,Box2D]==0.18.0 26 | # On Windows, install atari_py using: 27 | # pip install --no-index -f https://github.com/Kojoley/atari-py/releases atari_py 28 | 29 | ##### Image manipulation 30 | Pillow==8.1.2 31 | graphviz==0.16 32 | pyglet==1.5.0 33 | scikit-image==0.18.1 34 | 35 | #pyvirtualdisplay # needed in chapter 16, if on a headless server 36 | # (i.e., without screen, e.g., Colab or VM) 37 | 38 | 39 | ##### Additional utilities 40 | PrettyTable==2.1.0 41 | missingno==0.4.2 42 | 43 | # Efficient jobs (caching, parallelism, persistence) 44 | joblib==0.14.1 45 | 46 | # Nice utility to diff Jupyter Notebooks. 47 | nbdime==2.1.0 48 | 49 | # May be useful with Pandas for complex "where" clauses (e.g., Pandas 50 | # tutorial). 51 | numexpr==2.7.2 52 | 53 | 54 | nltk==3.5 55 | urlextract==1.2.0 56 | 57 | -------------------------------------------------------------------------------- /data/AUC.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/naikshubham/Predictive-Analytics-in-Python/5cc85ce6585cb3eb5a4253f168e5d46174af8083/data/AUC.JPG -------------------------------------------------------------------------------- /data/PIG_continuous.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/naikshubham/Predictive-Analytics-in-Python/5cc85ce6585cb3eb5a4253f168e5d46174af8083/data/PIG_continuous.JPG -------------------------------------------------------------------------------- /data/cum_gain.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/naikshubham/Predictive-Analytics-in-Python/5cc85ce6585cb3eb5a4253f168e5d46174af8083/data/cum_gain.JPG -------------------------------------------------------------------------------- /data/cumulative_gains.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/naikshubham/Predictive-Analytics-in-Python/5cc85ce6585cb3eb5a4253f168e5d46174af8083/data/cumulative_gains.JPG -------------------------------------------------------------------------------- /data/cutoff.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/naikshubham/Predictive-Analytics-in-Python/5cc85ce6585cb3eb5a4253f168e5d46174af8083/data/cutoff.JPG -------------------------------------------------------------------------------- /data/lift_curve.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/naikshubham/Predictive-Analytics-in-Python/5cc85ce6585cb3eb5a4253f168e5d46174af8083/data/lift_curve.JPG -------------------------------------------------------------------------------- /data/lift_curve_interpret.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/naikshubham/Predictive-Analytics-in-Python/5cc85ce6585cb3eb5a4253f168e5d46174af8083/data/lift_curve_interpret.JPG -------------------------------------------------------------------------------- /data/predictor_insights_graph.JPG: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/naikshubham/Predictive-Analytics-in-Python/5cc85ce6585cb3eb5a4253f168e5d46174af8083/data/predictor_insights_graph.JPG --------------------------------------------------------------------------------