├── .gitignore ├── .travis.yml ├── LICENSE ├── README.rst ├── setup.cfg ├── setup.py ├── statscounter ├── __init__.py ├── _stats.py ├── stats.py └── statscounter.py └── tests ├── test_stats.py └── test_statscounter.py /.gitignore: -------------------------------------------------------------------------------- 1 | # Byte-compiled / optimized / DLL files 2 | __pycache__/ 3 | *.py[cod] 4 | 5 | # C extensions 6 | *.so 7 | 8 | # Distribution / packaging 9 | .Python 10 | env/ 11 | build/ 12 | develop-eggs/ 13 | dist/ 14 | downloads/ 15 | eggs/ 16 | .eggs/ 17 | lib/ 18 | lib64/ 19 | parts/ 20 | sdist/ 21 | var/ 22 | *.egg-info/ 23 | .installed.cfg 24 | *.egg 25 | 26 | # PyInstaller 27 | # Usually these files are written by a python script from a template 28 | # before PyInstaller builds the exe, so as to inject date/other infos into it. 29 | *.manifest 30 | *.spec 31 | 32 | # Installer logs 33 | pip-log.txt 34 | pip-delete-this-directory.txt 35 | 36 | # Unit test / coverage reports 37 | htmlcov/ 38 | .tox/ 39 | .coverage 40 | .coverage.* 41 | .cache 42 | nosetests.xml 43 | coverage.xml 44 | *,cover 45 | 46 | # Translations 47 | *.mo 48 | *.pot 49 | 50 | # Django stuff: 51 | *.log 52 | 53 | # Sphinx documentation 54 | docs/_build/ 55 | 56 | # PyBuilder 57 | target/ 58 | -------------------------------------------------------------------------------- /.travis.yml: -------------------------------------------------------------------------------- 1 | language: python 2 | python: 3 | - "2.7" 4 | - "pypy" 5 | - "3.2" 6 | - "3.3" 7 | - "3.4" 8 | - "nightly" 9 | install: pip install . 10 | script: python setup.py test 11 | sudo: false 12 | notifications: 13 | email: false 14 | matrix: 15 | fast_finish: true 16 | allow_failures: 17 | - python: "nightly" 18 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | GNU GENERAL PUBLIC LICENSE 2 | Version 2, June 1991 3 | 4 | Copyright (C) 1989, 1991 Free Software Foundation, Inc., 5 | 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA 6 | Everyone is permitted to copy and distribute verbatim copies 7 | of this license document, but changing it is not allowed. 8 | 9 | Preamble 10 | 11 | The licenses for most software are designed to take away your 12 | freedom to share and change it. By contrast, the GNU General Public 13 | License is intended to guarantee your freedom to share and change free 14 | software--to make sure the software is free for all its users. This 15 | General Public License applies to most of the Free Software 16 | Foundation's software and to any other program whose authors commit to 17 | using it. (Some other Free Software Foundation software is covered by 18 | the GNU Lesser General Public License instead.) You can apply it to 19 | your programs, too. 20 | 21 | When we speak of free software, we are referring to freedom, not 22 | price. Our General Public Licenses are designed to make sure that you 23 | have the freedom to distribute copies of free software (and charge for 24 | this service if you wish), that you receive source code or can get it 25 | if you want it, that you can change the software or use pieces of it 26 | in new free programs; and that you know you can do these things. 27 | 28 | To protect your rights, we need to make restrictions that forbid 29 | anyone to deny you these rights or to ask you to surrender the rights. 30 | These restrictions translate to certain responsibilities for you if you 31 | distribute copies of the software, or if you modify it. 32 | 33 | For example, if you distribute copies of such a program, whether 34 | gratis or for a fee, you must give the recipients all the rights that 35 | you have. You must make sure that they, too, receive or can get the 36 | source code. And you must show them these terms so they know their 37 | rights. 38 | 39 | We protect your rights with two steps: (1) copyright the software, and 40 | (2) offer you this license which gives you legal permission to copy, 41 | distribute and/or modify the software. 42 | 43 | Also, for each author's protection and ours, we want to make certain 44 | that everyone understands that there is no warranty for this free 45 | software. If the software is modified by someone else and passed on, we 46 | want its recipients to know that what they have is not the original, so 47 | that any problems introduced by others will not reflect on the original 48 | authors' reputations. 49 | 50 | Finally, any free program is threatened constantly by software 51 | patents. We wish to avoid the danger that redistributors of a free 52 | program will individually obtain patent licenses, in effect making the 53 | program proprietary. To prevent this, we have made it clear that any 54 | patent must be licensed for everyone's free use or not licensed at all. 55 | 56 | The precise terms and conditions for copying, distribution and 57 | modification follow. 58 | 59 | GNU GENERAL PUBLIC LICENSE 60 | TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION 61 | 62 | 0. This License applies to any program or other work which contains 63 | a notice placed by the copyright holder saying it may be distributed 64 | under the terms of this General Public License. The "Program", below, 65 | refers to any such program or work, and a "work based on the Program" 66 | means either the Program or any derivative work under copyright law: 67 | that is to say, a work containing the Program or a portion of it, 68 | either verbatim or with modifications and/or translated into another 69 | language. (Hereinafter, translation is included without limitation in 70 | the term "modification".) Each licensee is addressed as "you". 71 | 72 | Activities other than copying, distribution and modification are not 73 | covered by this License; they are outside its scope. The act of 74 | running the Program is not restricted, and the output from the Program 75 | is covered only if its contents constitute a work based on the 76 | Program (independent of having been made by running the Program). 77 | Whether that is true depends on what the Program does. 78 | 79 | 1. You may copy and distribute verbatim copies of the Program's 80 | source code as you receive it, in any medium, provided that you 81 | conspicuously and appropriately publish on each copy an appropriate 82 | copyright notice and disclaimer of warranty; keep intact all the 83 | notices that refer to this License and to the absence of any warranty; 84 | and give any other recipients of the Program a copy of this License 85 | along with the Program. 86 | 87 | You may charge a fee for the physical act of transferring a copy, and 88 | you may at your option offer warranty protection in exchange for a fee. 89 | 90 | 2. You may modify your copy or copies of the Program or any portion 91 | of it, thus forming a work based on the Program, and copy and 92 | distribute such modifications or work under the terms of Section 1 93 | above, provided that you also meet all of these conditions: 94 | 95 | a) You must cause the modified files to carry prominent notices 96 | stating that you changed the files and the date of any change. 97 | 98 | b) You must cause any work that you distribute or publish, that in 99 | whole or in part contains or is derived from the Program or any 100 | part thereof, to be licensed as a whole at no charge to all third 101 | parties under the terms of this License. 102 | 103 | c) If the modified program normally reads commands interactively 104 | when run, you must cause it, when started running for such 105 | interactive use in the most ordinary way, to print or display an 106 | announcement including an appropriate copyright notice and a 107 | notice that there is no warranty (or else, saying that you provide 108 | a warranty) and that users may redistribute the program under 109 | these conditions, and telling the user how to view a copy of this 110 | License. (Exception: if the Program itself is interactive but 111 | does not normally print such an announcement, your work based on 112 | the Program is not required to print an announcement.) 113 | 114 | These requirements apply to the modified work as a whole. If 115 | identifiable sections of that work are not derived from the Program, 116 | and can be reasonably considered independent and separate works in 117 | themselves, then this License, and its terms, do not apply to those 118 | sections when you distribute them as separate works. But when you 119 | distribute the same sections as part of a whole which is a work based 120 | on the Program, the distribution of the whole must be on the terms of 121 | this License, whose permissions for other licensees extend to the 122 | entire whole, and thus to each and every part regardless of who wrote it. 123 | 124 | Thus, it is not the intent of this section to claim rights or contest 125 | your rights to work written entirely by you; rather, the intent is to 126 | exercise the right to control the distribution of derivative or 127 | collective works based on the Program. 128 | 129 | In addition, mere aggregation of another work not based on the Program 130 | with the Program (or with a work based on the Program) on a volume of 131 | a storage or distribution medium does not bring the other work under 132 | the scope of this License. 133 | 134 | 3. You may copy and distribute the Program (or a work based on it, 135 | under Section 2) in object code or executable form under the terms of 136 | Sections 1 and 2 above provided that you also do one of the following: 137 | 138 | a) Accompany it with the complete corresponding machine-readable 139 | source code, which must be distributed under the terms of Sections 140 | 1 and 2 above on a medium customarily used for software interchange; or, 141 | 142 | b) Accompany it with a written offer, valid for at least three 143 | years, to give any third party, for a charge no more than your 144 | cost of physically performing source distribution, a complete 145 | machine-readable copy of the corresponding source code, to be 146 | distributed under the terms of Sections 1 and 2 above on a medium 147 | customarily used for software interchange; or, 148 | 149 | c) Accompany it with the information you received as to the offer 150 | to distribute corresponding source code. (This alternative is 151 | allowed only for noncommercial distribution and only if you 152 | received the program in object code or executable form with such 153 | an offer, in accord with Subsection b above.) 154 | 155 | The source code for a work means the preferred form of the work for 156 | making modifications to it. For an executable work, complete source 157 | code means all the source code for all modules it contains, plus any 158 | associated interface definition files, plus the scripts used to 159 | control compilation and installation of the executable. However, as a 160 | special exception, the source code distributed need not include 161 | anything that is normally distributed (in either source or binary 162 | form) with the major components (compiler, kernel, and so on) of the 163 | operating system on which the executable runs, unless that component 164 | itself accompanies the executable. 165 | 166 | If distribution of executable or object code is made by offering 167 | access to copy from a designated place, then offering equivalent 168 | access to copy the source code from the same place counts as 169 | distribution of the source code, even though third parties are not 170 | compelled to copy the source along with the object code. 171 | 172 | 4. You may not copy, modify, sublicense, or distribute the Program 173 | except as expressly provided under this License. Any attempt 174 | otherwise to copy, modify, sublicense or distribute the Program is 175 | void, and will automatically terminate your rights under this License. 176 | However, parties who have received copies, or rights, from you under 177 | this License will not have their licenses terminated so long as such 178 | parties remain in full compliance. 179 | 180 | 5. You are not required to accept this License, since you have not 181 | signed it. However, nothing else grants you permission to modify or 182 | distribute the Program or its derivative works. These actions are 183 | prohibited by law if you do not accept this License. Therefore, by 184 | modifying or distributing the Program (or any work based on the 185 | Program), you indicate your acceptance of this License to do so, and 186 | all its terms and conditions for copying, distributing or modifying 187 | the Program or works based on it. 188 | 189 | 6. Each time you redistribute the Program (or any work based on the 190 | Program), the recipient automatically receives a license from the 191 | original licensor to copy, distribute or modify the Program subject to 192 | these terms and conditions. You may not impose any further 193 | restrictions on the recipients' exercise of the rights granted herein. 194 | You are not responsible for enforcing compliance by third parties to 195 | this License. 196 | 197 | 7. If, as a consequence of a court judgment or allegation of patent 198 | infringement or for any other reason (not limited to patent issues), 199 | conditions are imposed on you (whether by court order, agreement or 200 | otherwise) that contradict the conditions of this License, they do not 201 | excuse you from the conditions of this License. If you cannot 202 | distribute so as to satisfy simultaneously your obligations under this 203 | License and any other pertinent obligations, then as a consequence you 204 | may not distribute the Program at all. For example, if a patent 205 | license would not permit royalty-free redistribution of the Program by 206 | all those who receive copies directly or indirectly through you, then 207 | the only way you could satisfy both it and this License would be to 208 | refrain entirely from distribution of the Program. 209 | 210 | If any portion of this section is held invalid or unenforceable under 211 | any particular circumstance, the balance of the section is intended to 212 | apply and the section as a whole is intended to apply in other 213 | circumstances. 214 | 215 | It is not the purpose of this section to induce you to infringe any 216 | patents or other property right claims or to contest validity of any 217 | such claims; this section has the sole purpose of protecting the 218 | integrity of the free software distribution system, which is 219 | implemented by public license practices. Many people have made 220 | generous contributions to the wide range of software distributed 221 | through that system in reliance on consistent application of that 222 | system; it is up to the author/donor to decide if he or she is willing 223 | to distribute software through any other system and a licensee cannot 224 | impose that choice. 225 | 226 | This section is intended to make thoroughly clear what is believed to 227 | be a consequence of the rest of this License. 228 | 229 | 8. If the distribution and/or use of the Program is restricted in 230 | certain countries either by patents or by copyrighted interfaces, the 231 | original copyright holder who places the Program under this License 232 | may add an explicit geographical distribution limitation excluding 233 | those countries, so that distribution is permitted only in or among 234 | countries not thus excluded. In such case, this License incorporates 235 | the limitation as if written in the body of this License. 236 | 237 | 9. The Free Software Foundation may publish revised and/or new versions 238 | of the General Public License from time to time. Such new versions will 239 | be similar in spirit to the present version, but may differ in detail to 240 | address new problems or concerns. 241 | 242 | Each version is given a distinguishing version number. If the Program 243 | specifies a version number of this License which applies to it and "any 244 | later version", you have the option of following the terms and conditions 245 | either of that version or of any later version published by the Free 246 | Software Foundation. If the Program does not specify a version number of 247 | this License, you may choose any version ever published by the Free Software 248 | Foundation. 249 | 250 | 10. If you wish to incorporate parts of the Program into other free 251 | programs whose distribution conditions are different, write to the author 252 | to ask for permission. For software which is copyrighted by the Free 253 | Software Foundation, write to the Free Software Foundation; we sometimes 254 | make exceptions for this. Our decision will be guided by the two goals 255 | of preserving the free status of all derivatives of our free software and 256 | of promoting the sharing and reuse of software generally. 257 | 258 | NO WARRANTY 259 | 260 | 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY 261 | FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN 262 | OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES 263 | PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED 264 | OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF 265 | MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS 266 | TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE 267 | PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, 268 | REPAIR OR CORRECTION. 269 | 270 | 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING 271 | WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR 272 | REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, 273 | INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING 274 | OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED 275 | TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY 276 | YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER 277 | PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE 278 | POSSIBILITY OF SUCH DAMAGES. 279 | 280 | END OF TERMS AND CONDITIONS 281 | 282 | How to Apply These Terms to Your New Programs 283 | 284 | If you develop a new program, and you want it to be of the greatest 285 | possible use to the public, the best way to achieve this is to make it 286 | free software which everyone can redistribute and change under these terms. 287 | 288 | To do so, attach the following notices to the program. It is safest 289 | to attach them to the start of each source file to most effectively 290 | convey the exclusion of warranty; and each file should have at least 291 | the "copyright" line and a pointer to where the full notice is found. 292 | 293 | {description} 294 | Copyright (C) {year} {fullname} 295 | 296 | This program is free software; you can redistribute it and/or modify 297 | it under the terms of the GNU General Public License as published by 298 | the Free Software Foundation; either version 2 of the License, or 299 | (at your option) any later version. 300 | 301 | This program is distributed in the hope that it will be useful, 302 | but WITHOUT ANY WARRANTY; without even the implied warranty of 303 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 304 | GNU General Public License for more details. 305 | 306 | You should have received a copy of the GNU General Public License along 307 | with this program; if not, write to the Free Software Foundation, Inc., 308 | 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. 309 | 310 | Also add information on how to contact you by electronic and paper mail. 311 | 312 | If the program is interactive, make it output a short notice like this 313 | when it starts in an interactive mode: 314 | 315 | Gnomovision version 69, Copyright (C) year name of author 316 | Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'. 317 | This is free software, and you are welcome to redistribute it 318 | under certain conditions; type `show c' for details. 319 | 320 | The hypothetical commands `show w' and `show c' should show the appropriate 321 | parts of the General Public License. Of course, the commands you use may 322 | be called something other than `show w' and `show c'; they could even be 323 | mouse-clicks or menu items--whatever suits your program. 324 | 325 | You should also get your employer (if you work as a programmer) or your 326 | school, if any, to sign a "copyright disclaimer" for the program, if 327 | necessary. Here is a sample; alter the names: 328 | 329 | Yoyodyne, Inc., hereby disclaims all copyright interest in the program 330 | `Gnomovision' (which makes passes at compilers) written by James Hacker. 331 | 332 | {signature of Ty Coon}, 1 April 1989 333 | Ty Coon, President of Vice 334 | 335 | This General Public License does not permit incorporating your program into 336 | proprietary programs. If your program is a subroutine library, you may 337 | consider it more useful to permit linking proprietary applications with the 338 | library. If this is what you want to do, use the GNU Lesser General 339 | Public License instead of this License. 340 | 341 | -------------------------------------------------------------------------------- /README.rst: -------------------------------------------------------------------------------- 1 | StatsCounter: A statistics-enabled Python container 2 | --------------------------------------------------- 3 | 4 | :: 5 | 6 | _ _ _ 7 | ___| |_ __ _| |_ ___ ___ ___ _ _ _ __ | |_ ___ _ __ 8 | / __| __/ _` | __/ __|/ __/ _ \| | | | '_ \| __/ _ \ '__| 9 | \__ \ || (_| | |_\__ \ (_| (_) | |_| | | | | || __/ | 10 | |___/\__\__,_|\__|___/\___\___/ \__,_|_| |_|\__\___|_| 11 | 12 | 13 | StatsCounter is a GNU Licensed, statistics powered version 14 | of Python's standard library ``Counter`` class. It attaches 15 | several helpful methods that can be used to make your 16 | data-driven uses a breeze. 17 | 18 | Usage 19 | ----- 20 | 21 | As a histogram 22 | ~~~~~~~~~~~~~~ 23 | 24 | .. code-block:: python 25 | 26 | >>> import statscounter as stats 27 | >>> letter_freq = stats.StatsCounter(a=1, b=2, c=3, d=4, e=4, f=6) 28 | >>> letter_freq.mean() # average frequency 29 | 3.3333333333333335 30 | >>> letter_freq.mode() # most frequent element 31 | 4 32 | >>> letter_freq.median() # the median number (avg if even # of items) 33 | 3.5 34 | >>> letter_freq.variance() # sample variance 35 | 3.066666666666667 36 | >>> letter_freq.stdev() # sample standard deviation 37 | 1.7511900715418263 38 | >>> letter_freq.pvariance() # population variance 39 | 2.555555555555556 40 | >>> letter_freq.pstdev() # population std. dev. 41 | 1.5986105077709065 42 | >>> letter_freq.max() # the maximum value 43 | 6 44 | >>> letter_freq.argmax() # the argument yielding the maximum value 45 | "f" 46 | 47 | As a utility 48 | ~~~~~~~~~~~~ 49 | 50 | .. code-block:: python 51 | 52 | >>> import statscounter as stats 53 | >>> stats.mean([1, 2, 3, 4, 4, 6]) # average frequency 54 | 3.3333333333333335 55 | >>> stats.mode([1, 2, 3, 4, 4, 6]) # most frequent element 56 | 4 57 | >>> stats.median([1, 2, 3, 4, 4, 6]) # the median number (avg if even # of items) 58 | 3.5 59 | >>> stats.variance([1, 2, 3, 4, 4, 6]) # sample variance 60 | 3.066666666666667 61 | >>> stats.stdev([1, 2, 3, 4, 4, 6]) # sample standard deviation 62 | 1.7511900715418263 63 | >>> stats.pvariance([1, 2, 3, 4, 4, 6]) # population variance 64 | 2.555555555555556 65 | >>> stats.pstdev([1, 2, 3, 4, 4, 6]) # population std. dev. 66 | 1.5986105077709065 67 | -------------------------------------------------------------------------------- /setup.cfg: -------------------------------------------------------------------------------- 1 | [metadata] 2 | description-file = README.rst 3 | -------------------------------------------------------------------------------- /setup.py: -------------------------------------------------------------------------------- 1 | import sys 2 | from setuptools import setup 3 | from setuptools.command.test import test as TestCommand 4 | 5 | 6 | class PyTest(TestCommand): 7 | user_options = [] 8 | 9 | def initialize_options(self): 10 | TestCommand.initialize_options(self) 11 | self.pytest_args = [] 12 | 13 | def finalize_options(self): 14 | TestCommand.finalize_options(self) 15 | self.test_args = [] 16 | self.test_suite = True 17 | 18 | def run_tests(self): 19 | import pytest 20 | errno = pytest.main(self.pytest_args) 21 | sys.exit(errno) 22 | 23 | 24 | def readme(): 25 | with open('README.rst') as f: 26 | return f.read() 27 | 28 | 29 | setup( 30 | name='statscounter', 31 | version='0.0.010', 32 | url='https://github.com/datalib/statscounter', 33 | license='MIT', 34 | description="Python's missing statistical Swiss Army knife", 35 | packages=['statscounter'], 36 | include_package_data=True, 37 | zip_safe=False, 38 | platforms='any', 39 | install_requires=[], 40 | keywords='stats statistics statistic statistical measurements \ 41 | mean average avg standard deviation std dev stddev \ 42 | variance Counter StatsCounter collection measurement \ 43 | measure count sum elements items', 44 | author='Rodrigo Palacios', 45 | author_email='rodrigopala91@gmail.com', 46 | scripts=[], 47 | package_data={}, 48 | tests_require=['pytest'], 49 | cmdclass={'test': PyTest}, 50 | ) 51 | -------------------------------------------------------------------------------- /statscounter/__init__.py: -------------------------------------------------------------------------------- 1 | from .statscounter import StatsCounter 2 | -------------------------------------------------------------------------------- /statscounter/_stats.py: -------------------------------------------------------------------------------- 1 | """ 2 | This module is derived from the stats module available in Python 3.4 3 | 4 | https://hg.python.org/cpython/file/3.4/Lib/statistics.py 5 | """ 6 | 7 | from __future__ import division 8 | 9 | import collections 10 | import math 11 | from itertools import chain 12 | 13 | from fractions import Fraction as F 14 | from decimal import Decimal as D 15 | 16 | 17 | class StatisticsError(ValueError): 18 | pass 19 | 20 | # === Private utilities === # 21 | 22 | 23 | def _first(iterable): 24 | for item in iterable: 25 | return item 26 | 27 | 28 | def _sum(data): 29 | """_sum(data [, start]) -> value 30 | Return a high-precision sum of the given numeric data. If optional 31 | argument ``start`` is given, it is added to the total. If ``data`` is 32 | empty, ``start`` (defaulting to 0) is returned. 33 | Examples 34 | -------- 35 | >>> _sum([3, 2.25, 4.5, -0.5, 1.0], 0.75) 36 | 11.0 37 | Some sources of round-off error will be avoided: 38 | >>> _sum([1e50, 1, -1e50] * 1000) # Built-in sum returns zero. 39 | 1000.0 40 | Fractions and Decimals are also supported: 41 | >>> from fractions import Fraction as F 42 | >>> _sum([F(2, 3), F(7, 5), F(1, 4), F(5, 6)]) 43 | Fraction(63, 20) 44 | >>> from decimal import Decimal as D 45 | >>> data = [D("0.1375"), D("0.2108"), D("0.3061"), D("0.0419")] 46 | >>> _sum(data) 47 | Decimal('0.6963') 48 | Mixed types are currently treated as an error, except that int is 49 | allowed. 50 | """ 51 | data = iter(data) 52 | n = _first(data) 53 | 54 | if n is not None: 55 | data = chain([n], data) 56 | if isinstance(n, F): 57 | return math.fsum(data) 58 | return sum(data) 59 | return 0 60 | 61 | 62 | # === Measures of central tendency (averages) === 63 | 64 | def mean(data): 65 | """Return the sample arithmetic mean of data. 66 | 67 | >>> mean([1, 2, 3, 4, 4]) 68 | 2.8 69 | 70 | >>> from fractions import Fraction as F 71 | >>> mean([F(3, 7), F(1, 21), F(5, 3), F(1, 3)]) 72 | Fraction(13, 21) 73 | 74 | >>> from decimal import Decimal as D 75 | >>> mean([D("0.5"), D("0.75"), D("0.625"), D("0.375")]) 76 | Decimal('0.5625') 77 | 78 | If ``data`` is empty, StatisticsError will be raised. 79 | """ 80 | if iter(data) is data: 81 | data = list(data) 82 | n = len(data) 83 | if n < 1: 84 | raise StatisticsError('mean requires at least one data point') 85 | return _sum(data)/n 86 | 87 | 88 | # FIXME: investigate ways to calculate medians without sorting? Quickselect? 89 | def median(data): 90 | """Return the median (middle value) of numeric data. 91 | 92 | When the number of data points is odd, return the middle data point. 93 | When the number of data points is even, the median is interpolated by 94 | taking the average of the two middle values: 95 | 96 | >>> median([1, 3, 5]) 97 | 3 98 | >>> median([1, 3, 5, 7]) 99 | 4.0 100 | 101 | """ 102 | data = sorted(data) 103 | n = len(data) 104 | if n == 0: 105 | raise StatisticsError("no median for empty data") 106 | if n % 2: 107 | return data[n//2] 108 | else: 109 | i = n//2 110 | return (data[i - 1] + data[i])/2 111 | 112 | 113 | def median_low(data): 114 | """Return the low median of numeric data. 115 | 116 | When the number of data points is odd, the middle value is returned. 117 | When it is even, the smaller of the two middle values is returned. 118 | 119 | >>> median_low([1, 3, 5]) 120 | 3 121 | >>> median_low([1, 3, 5, 7]) 122 | 3 123 | 124 | """ 125 | data = sorted(data) 126 | n = len(data) 127 | if n == 0: 128 | raise StatisticsError("no median for empty data") 129 | if n%2 == 1: 130 | return data[n//2] 131 | else: 132 | return data[n//2 - 1] 133 | 134 | 135 | def median_high(data): 136 | """Return the high median of data. 137 | 138 | When the number of data points is odd, the middle value is returned. 139 | When it is even, the larger of the two middle values is returned. 140 | 141 | >>> median_high([1, 3, 5]) 142 | 3 143 | >>> median_high([1, 3, 5, 7]) 144 | 5 145 | 146 | """ 147 | data = sorted(data) 148 | n = len(data) 149 | if n == 0: 150 | raise StatisticsError("no median for empty data") 151 | return data[n//2] 152 | 153 | 154 | def median_grouped(data, interval=1): 155 | """"Return the 50th percentile (median) of grouped continuous data. 156 | 157 | >>> median_grouped([1, 2, 2, 3, 4, 4, 4, 4, 4, 5]) 158 | 3.7 159 | >>> median_grouped([52, 52, 53, 54]) 160 | 52.5 161 | 162 | This calculates the median as the 50th percentile, and should be 163 | used when your data is continuous and grouped. In the above example, 164 | the values 1, 2, 3, etc. actually represent the midpoint of classes 165 | 0.5-1.5, 1.5-2.5, 2.5-3.5, etc. The middle value falls somewhere in 166 | class 3.5-4.5, and interpolation is used to estimate it. 167 | 168 | Optional argument ``interval`` represents the class interval, and 169 | defaults to 1. Changing the class interval naturally will change the 170 | interpolated 50th percentile value: 171 | 172 | >>> median_grouped([1, 3, 3, 5, 7], interval=1) 173 | 3.25 174 | >>> median_grouped([1, 3, 3, 5, 7], interval=2) 175 | 3.5 176 | 177 | This function does not check whether the data points are at least 178 | ``interval`` apart. 179 | """ 180 | data = sorted(data) 181 | n = len(data) 182 | if n == 0: 183 | raise StatisticsError("no median for empty data") 184 | elif n == 1: 185 | return data[0] 186 | # Find the value at the midpoint. Remember this corresponds to the 187 | # centre of the class interval. 188 | x = data[n//2] 189 | for obj in (x, interval): 190 | if isinstance(obj, (str, bytes)): 191 | raise TypeError('expected number but got %r' % obj) 192 | try: 193 | L = x - interval/2 # The lower limit of the median interval. 194 | except TypeError: 195 | # Mixed type. For now we just coerce to float. 196 | L = float(x) - float(interval)/2 197 | cf = data.index(x) # Number of values below the median interval. 198 | # FIXME The following line could be more efficient for big lists. 199 | f = data.count(x) # Number of data points in the median interval. 200 | return L + interval*(n/2 - cf)/f 201 | 202 | 203 | def mode(data): 204 | """Return the most common data point from discrete or nominal data. 205 | 206 | ``mode`` assumes discrete data, and returns a single value. This is the 207 | standard treatment of the mode as commonly taught in schools: 208 | 209 | >>> mode([1, 1, 2, 3, 3, 3, 3, 4]) 210 | 3 211 | 212 | This also works with nominal (non-numeric) data: 213 | 214 | >>> mode(["red", "blue", "blue", "red", "green", "red", "red"]) 215 | 'red' 216 | """ 217 | 218 | # Generate a table of sorted (value, frequency) pairs. 219 | hist = collections.Counter(data) 220 | top = hist.most_common(2) 221 | 222 | if len(top) == 1: 223 | return top[0][0] 224 | elif not top: 225 | raise StatisticsError('no mode for empty data') 226 | elif top[0][1] == top[1][1]: 227 | raise StatisticsError( 228 | 'no unique mode; found %d equally common values' % len(hist) 229 | ) 230 | else: 231 | return top[0][0] 232 | 233 | 234 | # === Measures of spread === 235 | 236 | # See http://mathworld.wolfram.com/Variance.html 237 | # http://mathworld.wolfram.com/SampleVariance.html 238 | # http://en.wikipedia.org/wiki/Algorithms_for_calculating_variance 239 | # 240 | # Under no circumstances use the so-called "computational formula for 241 | # variance", as that is only suitable for hand calculations with a small 242 | # amount of low-precision data. It has terrible numeric properties. 243 | # 244 | # See a comparison of three computational methods here: 245 | # http://www.johndcook.com/blog/2008/09/26/comparing-three-methods-of-computing-standard-deviation/ 246 | 247 | def _ss(data, c=None): 248 | """Return sum of square deviations of sequence data. 249 | 250 | If ``c`` is None, the mean is calculated in one pass, and the deviations 251 | from the mean are calculated in a second pass. Otherwise, deviations are 252 | calculated from ``c`` as given. Use the second case with care, as it can 253 | lead to garbage results. 254 | """ 255 | if c is None: 256 | c = mean(data) 257 | #print(data) 258 | ss = _sum((x-c)**2 for x in data) 259 | # The following sum should mathematically equal zero, but due to rounding 260 | # error may not. 261 | ss -= _sum((x-c) for x in data)**2/len(data) 262 | assert not ss < 0, 'negative sum of square deviations: %f' % ss 263 | return ss 264 | 265 | 266 | def variance(data, xbar=None): 267 | """Return the sample variance of data. 268 | 269 | data should be an iterable of Real-valued numbers, with at least two 270 | values. The optional argument xbar, if given, should be the mean of 271 | the data. If it is missing or None, the mean is automatically calculated. 272 | 273 | Use this function when your data is a sample from a population. To 274 | calculate the variance from the entire population, see ``pvariance``. 275 | 276 | Examples: 277 | 278 | >>> data = [2.75, 1.75, 1.25, 0.25, 0.5, 1.25, 3.5] 279 | >>> variance(data) 280 | 1.3720238095238095 281 | 282 | If you have already calculated the mean of your data, you can pass it as 283 | the optional second argument ``xbar`` to avoid recalculating it: 284 | 285 | >>> m = mean(data) 286 | >>> variance(data, m) 287 | 1.3720238095238095 288 | 289 | This function does not check that ``xbar`` is actually the mean of 290 | ``data``. Giving arbitrary values for ``xbar`` may lead to invalid or 291 | impossible results. 292 | 293 | Decimals and Fractions are supported: 294 | 295 | >>> from decimal import Decimal as D 296 | >>> variance([D("27.5"), D("30.25"), D("30.25"), D("34.5"), D("41.75")]) 297 | Decimal('31.01875') 298 | 299 | >>> from fractions import Fraction as F 300 | >>> variance([F(1, 6), F(1, 2), F(5, 3)]) 301 | Fraction(67, 108) 302 | """ 303 | if iter(data) is data: 304 | data = list(data) 305 | n = len(data) 306 | if n < 2: 307 | raise StatisticsError('variance requires at least two data points') 308 | ss = _ss(data, xbar) 309 | return ss/(n-1) 310 | 311 | 312 | def pvariance(data, mu=None): 313 | """Return the population variance of ``data``. 314 | 315 | data should be an iterable of Real-valued numbers, with at least one 316 | value. The optional argument mu, if given, should be the mean of 317 | the data. If it is missing or None, the mean is automatically calculated. 318 | 319 | Use this function to calculate the variance from the entire population. 320 | To estimate the variance from a sample, the ``variance`` function is 321 | usually a better choice. 322 | 323 | Examples: 324 | 325 | >>> data = [0.0, 0.25, 0.25, 1.25, 1.5, 1.75, 2.75, 3.25] 326 | >>> pvariance(data) 327 | 1.25 328 | 329 | If you have already calculated the mean of the data, you can pass it as 330 | the optional second argument to avoid recalculating it: 331 | 332 | >>> mu = mean(data) 333 | >>> pvariance(data, mu) 334 | 1.25 335 | 336 | This function does not check that ``mu`` is actually the mean of ``data``. 337 | Giving arbitrary values for ``mu`` may lead to invalid or impossible 338 | results. 339 | 340 | Decimals and Fractions are supported: 341 | 342 | >>> from decimal import Decimal as D 343 | >>> pvariance([D("27.5"), D("30.25"), D("30.25"), D("34.5"), D("41.75")]) 344 | Decimal('24.815') 345 | 346 | >>> from fractions import Fraction as F 347 | >>> pvariance([F(1, 4), F(5, 4), F(1, 2)]) 348 | Fraction(13, 72) 349 | 350 | """ 351 | if iter(data) is data: 352 | data = list(data) 353 | n = len(data) 354 | if n < 1: 355 | raise StatisticsError('pvariance requires at least one data point') 356 | ss = _ss(data, mu) 357 | return ss/n 358 | 359 | 360 | def stdev(data, xbar=None): 361 | """Return the square root of the sample variance. 362 | 363 | See ``variance`` for arguments and other details. 364 | 365 | >>> stdev([1.5, 2.5, 2.5, 2.75, 3.25, 4.75]) 366 | 1.0810874155219827 367 | 368 | """ 369 | var = variance(data, xbar) 370 | try: 371 | return var.sqrt() 372 | except AttributeError: 373 | return math.sqrt(var) 374 | 375 | 376 | def pstdev(data, mu=None): 377 | """Return the square root of the population variance. 378 | 379 | See ``pvariance`` for arguments and other details. 380 | 381 | >>> pstdev([1.5, 2.5, 2.5, 2.75, 3.25, 4.75]) 382 | 0.986893273527251 383 | 384 | """ 385 | var = pvariance(data, mu) 386 | try: 387 | return var.sqrt() 388 | except AttributeError: 389 | return math.sqrt(var) 390 | -------------------------------------------------------------------------------- /statscounter/stats.py: -------------------------------------------------------------------------------- 1 | try: 2 | from statistics import * 3 | except ImportError: 4 | from ._stats import * 5 | -------------------------------------------------------------------------------- /statscounter/statscounter.py: -------------------------------------------------------------------------------- 1 | from __future__ import absolute_import 2 | """StatsCounter 3 | 4 | This module is derived from the stats module available 5 | in Python 3.4: 6 | 7 | https://hg.python.org/cpython/file/3.4/Lib/statistics.py 8 | 9 | Many a times I found myself wanting to to run simple 10 | averaging, summation, variance-calculating methods on the 11 | wonderful built-in collections.Counter class, and I would 12 | always include a those "helper" functions that allowed 13 | me to do just that. 14 | 15 | After the n-th time of doing the above mentioned ritual, 16 | I decided to look at the statistics module in Python3, 17 | and I was surprised to see that most of the code was 18 | written in Python that can be easily back-ported. 19 | 20 | Cheers, 21 | Rodrigo 22 | """ 23 | 24 | from collections import Counter 25 | import statscounter.stats as stats 26 | 27 | 28 | class StatsCounter(Counter): 29 | def mean(self): 30 | """ 31 | """ 32 | return stats.mean(self.values()) 33 | 34 | def median(self, ): 35 | """ 36 | """ 37 | return stats.median(self.values()) 38 | 39 | def median_low(self): 40 | """ 41 | """ 42 | return stats.median_low(self.values()) 43 | 44 | def median_high(self): 45 | """ 46 | """ 47 | return stats.median_high(self.values()) 48 | 49 | def median_grouped(self): 50 | """ 51 | """ 52 | return stats.median_grouped(self.values()) 53 | 54 | def mode(self): 55 | """ 56 | """ 57 | return stats.mode(self.values()) 58 | 59 | def variance(self): 60 | """ 61 | """ 62 | return stats.variance(self.values()) 63 | 64 | def pvariance(self): 65 | """ 66 | """ 67 | return stats.pvariance(self.values()) 68 | 69 | def stdev(self, ): 70 | """ 71 | """ 72 | return stats.stdev(self.values()) 73 | 74 | def pstdev(self): 75 | """ 76 | """ 77 | return stats.pstdev(self.values()) 78 | 79 | def best_pair(self): 80 | return self.most_common(1)[0] 81 | 82 | def argmax(self): 83 | """ 84 | """ 85 | key, _ = self.best_pair() 86 | return key 87 | 88 | def max(self): 89 | """ 90 | """ 91 | _, value = self.best_pair() 92 | return value 93 | 94 | def normalize(self): 95 | """ 96 | Sum the values in a Counter, then create a new Counter 97 | where each new value (while keeping the original key) 98 | is equal to the original value divided by sum of all the 99 | original values (this is sometimes referred to as the 100 | normalization constant). 101 | https://en.wikipedia.org/wiki/Normalization_(statistics) 102 | """ 103 | total = sum(self.values()) 104 | stats = {k: (v / float(total)) for k, v in self.items()} 105 | return StatsCounter(stats) 106 | 107 | def get_weighted_random_value(self): 108 | """ 109 | This will generate a value by creating a cumulative distribution, 110 | and a random number, and selecting the value who's cumulative 111 | distribution interval contains the generated random number. 112 | 113 | For example, if there's 0.7 chance of generating the letter "a" 114 | and 0.3 chance of generating the letter "b", then if you were to 115 | pick one letter 100 times over, the number of a's and b's you 116 | would have are likely to be around 70 and 30 respectively. 117 | 118 | The mechanics are known as "Cumulative distribution functions" 119 | (https://en.wikipedia.org/wiki/Cumulative_distribution_function) 120 | """ 121 | from bisect import bisect 122 | from random import random 123 | #http://stackoverflow.com/questions/4437250/choose-list-variable-given-probability-of-each-variable 124 | 125 | total = sum(self.values()) 126 | 127 | P = [(k, (v / float(total))) for k, v in self.items()] 128 | 129 | cdf = [P[0][1]] 130 | for i in range(1, len(P)): 131 | cdf.append(cdf[-1] + P[i][1]) 132 | 133 | return P[bisect(cdf, random())][0] 134 | 135 | 136 | def transform(self, key): 137 | """ 138 | """ 139 | dist = self 140 | newdist = StatsCounter() 141 | 142 | for k, v in dist.items(): 143 | newdist[key(k, v)] += v 144 | 145 | return newdist -------------------------------------------------------------------------------- /tests/test_stats.py: -------------------------------------------------------------------------------- 1 | from __future__ import division 2 | from statscounter.stats import mean, median, median_low, \ 3 | median_high, median_grouped, mode, \ 4 | stdev, pstdev, variance, pvariance 5 | 6 | 7 | def frange(x, y, jump): 8 | while x < y: 9 | yield x 10 | x += jump 11 | 12 | 13 | class TestStats: 14 | ints = list(range(10000)) 15 | floats = frange(0, 1, 0.001) 16 | 17 | def test_mean(self): 18 | m = mean(self.ints) 19 | d = (49995000)/10000 20 | assert m == d 21 | 22 | def test_median(self): 23 | m = median(self.ints) 24 | assert m == 4999.5 25 | 26 | def test_mode(self): 27 | m = mode(self.ints + [1]) 28 | assert m == 1 29 | 30 | def test_median_low(self): 31 | m = median_low(self.ints) 32 | assert m == 4999 33 | 34 | def test_median_high(self, ): 35 | m = median_high(self.ints) 36 | assert m == 5000 37 | 38 | def test_median_grouped(self, ): 39 | m = median_grouped(self.ints) 40 | assert m == 4999.5 41 | 42 | def test_variance(self): 43 | m = variance(self.ints) 44 | assert m == 8334166.666666667 45 | 46 | def test_stdev(self): 47 | m = stdev(self.ints) 48 | assert m == 2886.8956799071675 49 | 50 | def test_pvariance(self): 51 | m = pvariance(self.ints) 52 | assert m == 8333333.25 53 | 54 | def test_pstdev(self): 55 | m = pstdev(self.ints) 56 | assert m == 2886.751331514372 57 | -------------------------------------------------------------------------------- /tests/test_statscounter.py: -------------------------------------------------------------------------------- 1 | from __future__ import division 2 | from pytest import raises 3 | from statscounter import StatsCounter, stats 4 | 5 | 6 | class TestStatsCounter: 7 | counter_ints = StatsCounter({str(s):s for s in range(1000)}) 8 | 9 | def test_mean_int(self): 10 | m = self.counter_ints.mean() 11 | d = 499500/1000 12 | assert m == d 13 | 14 | def test_median_low(self): 15 | m = self.counter_ints.median_low() 16 | assert m == 499 17 | 18 | def test_median_high(self, ): 19 | m = self.counter_ints.median_high() 20 | assert m == 500 21 | 22 | def test_median_grouped(self, ): 23 | m = self.counter_ints.median_grouped() 24 | assert m == 499.5 25 | 26 | def test_mode(self): 27 | with raises(stats.StatisticsError): 28 | self.counter_ints.mode() 29 | 30 | def test_variance(self): 31 | m = self.counter_ints.variance() 32 | assert m == 83416.66666666667 33 | 34 | def test_stdev(self, ): 35 | m = self.counter_ints.stdev() 36 | assert m == 288.8194360957494 37 | 38 | def test_pvariance(self): 39 | m = self.counter_ints.pvariance() 40 | assert m == 83333.25 41 | 42 | def test_pstdev(self, ): 43 | m = self.counter_ints.pstdev() 44 | assert m == 288.6749902572095 45 | 46 | def test_argmax(self): 47 | m = self.counter_ints.argmax() 48 | assert m == '999' 49 | 50 | def test_max(self): 51 | m = self.counter_ints.max() 52 | assert m == 999 53 | 54 | def test_normalize(self): 55 | pdist = StatsCounter({1: 1, 2: 2, 3: 1}).normalize() 56 | assert pdist == { 57 | 1: 0.25, 58 | 2: 0.50, 59 | 3: 0.25, 60 | } 61 | 62 | def test_get_weighted_random_value(self): 63 | wrv = StatsCounter(a=10, b=3).get_weighted_random_value() 64 | assert wrv == "a" or "b" 65 | 66 | def test_transform(self): 67 | dist = StatsCounter({ 68 | 'of': 0.20, 69 | 'the': 0.50, 70 | 'that': 0.10, 71 | 'from': 0.20 72 | }) 73 | 74 | dist = dist.transform(lambda word, prob: word.startswith('t')) 75 | 76 | assert dist == StatsCounter({True: 0.6, False: 0.4}) --------------------------------------------------------------------------------