Full Code of milcent/benford_py for AI

master 0126c606ae9c cached

60 files

2.2 MB

584.6k tokens

272 symbols

1 requests

Download .txt

Showing preview only (2,339K chars total). Download the full file or copy to clipboard to get everything.

Repository: milcent/benford_py
Branch: master
Commit: 0126c606ae9c
Files: 60
Total size: 2.2 MB

Directory structure:
gitextract_ip3h_u68/

├── .github/
│   └── workflows/
│       ├── pylint.yml
│       └── python-package.yml
├── .gitignore
├── .pylintrc
├── .readthedocs.yml
├── CITATION.cff
├── Demo.ipynb
├── LICENSE.txt
├── MANIFEST.in
├── README-pypi.md
├── README.md
├── __init__.py
├── benford/
│   ├── __init__.py
│   ├── benford.py
│   ├── checks.py
│   ├── constants.py
│   ├── expected.py
│   ├── reports.py
│   ├── stats.py
│   ├── utils.py
│   └── viz.py
├── data/
│   └── SPY.csv
├── docs/
│   ├── Makefile
│   ├── build/
│   │   ├── doctrees/
│   │   │   ├── api.doctree
│   │   │   ├── benford.doctree
│   │   │   ├── environment.pickle
│   │   │   ├── index.doctree
│   │   │   └── modules.doctree
│   │   └── html/
│   │       ├── .buildinfo
│   │       ├── _modules/
│   │       │   ├── benford/
│   │       │   │   ├── benford.html
│   │       │   │   ├── expected.html
│   │       │   │   ├── stats.html
│   │       │   │   ├── utils.html
│   │       │   │   └── viz.html
│   │       │   └── index.html
│   │       ├── _sources/
│   │       │   ├── api.rst.txt
│   │       │   ├── index.rst.txt
│   │       │   └── modules.rst.txt
│   │       ├── api.html
│   │       ├── genindex.html
│   │       ├── index.html
│   │       ├── modules.html
│   │       ├── objects.inv
│   │       ├── py-modindex.html
│   │       ├── search.html
│   │       └── searchindex.js
│   ├── make.bat
│   ├── requirements.txt
│   └── source/
│       ├── api.rst
│       ├── conf.py
│       ├── index.rst
│       └── modules.rst
├── setup.cfg
├── setup.py
└── tests/
    ├── __init__.py
    ├── conftest.py
    ├── test_checks.py
    ├── test_expected.py
    ├── test_stats.py
    └── test_utils.py

================================================
FILE CONTENTS
================================================

================================================
FILE: .github/workflows/pylint.yml
================================================
name: Pylint

on: [push]

jobs:
  build:

    runs-on: ubuntu-latest

    steps:
      - uses: actions/checkout@v2
      - name: Set up Python 3.8
        uses: actions/setup-python@v1
        with:
          python-version: 3.8
      - name: Install dependencies
        run: |
          python -m pip install --upgrade pip
          pip install pylint numpy pandas matplotlib
          if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
      - name: Analysing the code with pylint
        run: |
          pylint `ls -R|grep .py$|xargs`


================================================
FILE: .github/workflows/python-package.yml
================================================
# This workflow will install Python dependencies, run tests and lint with a variety of Python versions
# For more information see: https://help.github.com/actions/language-and-framework-guides/using-python-with-github-actions

name: benford_py

on:
  push:
    branches: [ develop ]
  pull_request:
    branches: [ develop ]

jobs:
  build:

    runs-on: ubuntu-latest
    strategy:
      matrix:
        python-version: [3.6, 3.7, 3.8, 3.9]

    steps:
    - uses: actions/checkout@v2
    - name: Set up Python ${{ matrix.python-version }}
      uses: actions/setup-python@v2
      with:
        python-version: ${{ matrix.python-version }}
    - name: Install dependencies
      run: |
        python -m pip install --upgrade pip
        pip install flake8 pytest numpy pandas matplotlib
        if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
    - name: Lint with flake8
      run: |
        # stop the build if there are Python syntax errors or undefined names
        flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
        # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
        flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
    - name: Test with pytest
      run: |
        pytest


================================================
FILE: .gitignore
================================================
# Compiled python modules.
*.pyc

__pycache__/

# ipython notebook checkpoints
*.ipynb_checkpoints

# text editor backups
*~

# VS Code
.vscode/

# Jupyter NB Checkpoints
.ipynb_checkpoints/

# Setuptools distribution folder.
/dist/
/build/

# Python egg metadata, regenerated from source files by setuptools.
/*.egg-info

# Sphinx docs rendered files
# /docs/build/
_build
_static
_templates

# pytest
.pytest_cache/
#VSCode
.vscode/


================================================
FILE: .pylintrc
================================================
[MASTER]
disable=
    F0001, # No module named XXXXXXXX



ignored-classes=SQLObject,Registrant,scoped_session


================================================
FILE: .readthedocs.yml
================================================
# .readthedocs.yml
# Read the Docs configuration file
# See https://docs.readthedocs.io/en/stable/config-file/v2.html for details

# Required
version: 2

# Build documentation in the docs/ directory with Sphinx
sphinx:
  configuration: docs/source/conf.py

# Build documentation with MkDocs
#mkdocs:
#  configuration: mkdocs.yml

# Optionally build your docs in additional formats such as PDF and ePub
formats: all

# Optionally set the version of Python and requirements required to build your docs
python:
  version: 3.7
  install:
    - requirements: docs/requirements.txt

================================================
FILE: CITATION.cff
================================================
cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
- family-names: "Milcent"
  given-names: "Marcel"
  orcid: 
title: "Benford_py: a Python Implementation of Benford's Law Tests"
version: 0.5.0
doi: 
date-released: 2017
url: "https://github.com/milcent/benford_py"

================================================
FILE: Demo.ipynb
================================================
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Benford  for Python"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Current version: 0.4.3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Installation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Benford for python is a Package in PyPi, so you can install with *pip*:\n",
    "##### `$ pip install benford_py`\n",
    "### or\n",
    "##### `$ pip install benford-py`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Or you can cd into the site-packages subfolder of your python distribution (or environment) and clone from there:\n",
    "##### `$ git clone http://github.com/milcent/benford_py.git`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## [Documentation](https://benford-py.readthedocs.io)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Demo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### This demo assumes you have (at least) some familiarity with Benford's Law."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### First let's import some libraries and the benford module."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import benford as bf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Quick start"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Getting some public data with pandas, the S&P500 EFT quotes, up until Dec 2016.\n",
    "##### I have downloaded it and saved it in the data folder, so you should have it if you cloned the repo. Alternatively, you an downolad it [here](https://github.com/milcent/benford_py/blob/master/data/SPY.csv)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "sp = pd.read_csv('data/SPY.csv', index_col='Date', parse_dates=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Creating simple and log return  columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Open</th>\n",
       "      <th>High</th>\n",
       "      <th>Low</th>\n",
       "      <th>Close</th>\n",
       "      <th>Volume</th>\n",
       "      <th>Adj_Close</th>\n",
       "      <th>p_r</th>\n",
       "      <th>l_r</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-12-23</th>\n",
       "      <td>225.429993</td>\n",
       "      <td>225.720001</td>\n",
       "      <td>225.210007</td>\n",
       "      <td>225.710007</td>\n",
       "      <td>36251400</td>\n",
       "      <td>225.710007</td>\n",
       "      <td>0.001464</td>\n",
       "      <td>0.001463</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-27</th>\n",
       "      <td>226.020004</td>\n",
       "      <td>226.729996</td>\n",
       "      <td>226.000000</td>\n",
       "      <td>226.270004</td>\n",
       "      <td>41054400</td>\n",
       "      <td>226.270004</td>\n",
       "      <td>0.002481</td>\n",
       "      <td>0.002478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-28</th>\n",
       "      <td>226.570007</td>\n",
       "      <td>226.589996</td>\n",
       "      <td>224.270004</td>\n",
       "      <td>224.399994</td>\n",
       "      <td>59776300</td>\n",
       "      <td>224.399994</td>\n",
       "      <td>-0.008265</td>\n",
       "      <td>-0.008299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-29</th>\n",
       "      <td>224.479996</td>\n",
       "      <td>224.889999</td>\n",
       "      <td>223.839996</td>\n",
       "      <td>224.350006</td>\n",
       "      <td>47719500</td>\n",
       "      <td>224.350006</td>\n",
       "      <td>-0.000223</td>\n",
       "      <td>-0.000223</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-30</th>\n",
       "      <td>224.729996</td>\n",
       "      <td>224.830002</td>\n",
       "      <td>222.729996</td>\n",
       "      <td>223.529999</td>\n",
       "      <td>101301800</td>\n",
       "      <td>223.529999</td>\n",
       "      <td>-0.003655</td>\n",
       "      <td>-0.003662</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  Open        High         Low       Close     Volume  \\\n",
       "Date                                                                    \n",
       "2016-12-23  225.429993  225.720001  225.210007  225.710007   36251400   \n",
       "2016-12-27  226.020004  226.729996  226.000000  226.270004   41054400   \n",
       "2016-12-28  226.570007  226.589996  224.270004  224.399994   59776300   \n",
       "2016-12-29  224.479996  224.889999  223.839996  224.350006   47719500   \n",
       "2016-12-30  224.729996  224.830002  222.729996  223.529999  101301800   \n",
       "\n",
       "             Adj_Close       p_r       l_r  \n",
       "Date                                        \n",
       "2016-12-23  225.710007  0.001464  0.001463  \n",
       "2016-12-27  226.270004  0.002481  0.002478  \n",
       "2016-12-28  224.399994 -0.008265 -0.008299  \n",
       "2016-12-29  224.350006 -0.000223 -0.000223  \n",
       "2016-12-30  223.529999 -0.003655 -0.003662  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#adding '_' to facilitate handling the column\n",
    "sp.rename(columns={'Adj Close':'Adj_Close'}, inplace=True) \n",
    "sp['p_r'] = sp.Close/sp.Close.shift()-1        #simple returns\n",
    "sp['l_r'] = np.log(sp.Close/sp.Close.shift())  #log returns\n",
    "sp.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### First Digits Test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Let us see if the SPY log returns look like Benford.\n",
    "#### The `first_digits` function's first argument is the data to be analysed, which may be a pandas Series or a numpy 1D array.\n",
    "#### The `digs` argument tells the function which test to run: 1 for the _first_, 2 for the _first two_, and 3 for the _first three_ digits.The `decimals` parameter tells the function how many decimal places to consider when pre-processing the data. It defaults to 2, for dealing with currency, but since here we are dealing with tiny numbers, we will go with 8."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Initialized sequence with 5968 registries.\n",
      "\n",
      "Test performed on 5968 registries.\n",
      "Discarded 0 records < 1 after preparation.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f1d = bf.first_digits(sp.l_r, digs=1, decimals=8) # digs=1 for the first digit (1-9)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### The *first_digits* function draws the plot (default) with blue bars fot the digits found frequencies and a red line corresponding to the expected Benford proportions. \n",
    "#### It also returns a pandas DataFrame object with Counts, Found proportions and Expected values for each digit in the data studied.\n",
    "#### Zeros and NANs are dropped before processing.\n",
    "#### By default, (`verbose`=True) it also gives informaiton on the sample size and on the number of records eventually discarded during pre-processing (more on this later) ."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Counts</th>\n",
       "      <th>Found</th>\n",
       "      <th>Expected</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>First_1_Dig</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1835</td>\n",
       "      <td>0.307473</td>\n",
       "      <td>0.301030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>949</td>\n",
       "      <td>0.159015</td>\n",
       "      <td>0.176091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>668</td>\n",
       "      <td>0.111930</td>\n",
       "      <td>0.124939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>534</td>\n",
       "      <td>0.089477</td>\n",
       "      <td>0.096910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>494</td>\n",
       "      <td>0.082775</td>\n",
       "      <td>0.079181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>447</td>\n",
       "      <td>0.074899</td>\n",
       "      <td>0.066947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>386</td>\n",
       "      <td>0.064678</td>\n",
       "      <td>0.057992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>345</td>\n",
       "      <td>0.057808</td>\n",
       "      <td>0.051153</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>310</td>\n",
       "      <td>0.051944</td>\n",
       "      <td>0.045757</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Counts     Found  Expected\n",
       "First_1_Dig                            \n",
       "1              1835  0.307473  0.301030\n",
       "2               949  0.159015  0.176091\n",
       "3               668  0.111930  0.124939\n",
       "4               534  0.089477  0.096910\n",
       "5               494  0.082775  0.079181\n",
       "6               447  0.074899  0.066947\n",
       "7               386  0.064678  0.057992\n",
       "8               345  0.057808  0.051153\n",
       "9               310  0.051944  0.045757"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1d"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### First Two Digists"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Initialized sequence with 5968 registries.\n",
      "\n",
      "Test performed on 5968 registries.\n",
      "Discarded 0 records < 10 after preparation.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x972 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f2d = bf.first_digits(sp.l_r, digs=2, decimals=8) # Note the parameter digs=2!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### The variable `f2d` is assigned a pandas DataFrame with the Counts of registries with the digitss of interest, in this case 10-99 (First Two), the Found proportions and the Expected, Benfors ones, as shown bellow."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Counts</th>\n",
       "      <th>Found</th>\n",
       "      <th>Expected</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>First_2_Dig</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>298</td>\n",
       "      <td>0.049933</td>\n",
       "      <td>0.041393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>232</td>\n",
       "      <td>0.038874</td>\n",
       "      <td>0.037789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>203</td>\n",
       "      <td>0.034015</td>\n",
       "      <td>0.034762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>238</td>\n",
       "      <td>0.039879</td>\n",
       "      <td>0.032185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>178</td>\n",
       "      <td>0.029826</td>\n",
       "      <td>0.029963</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Counts     Found  Expected\n",
       "First_2_Dig                            \n",
       "10              298  0.049933  0.041393\n",
       "11              232  0.038874  0.037789\n",
       "12              203  0.034015  0.034762\n",
       "13              238  0.039879  0.032185\n",
       "14              178  0.029826  0.029963"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f2d.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Counts</th>\n",
       "      <th>Found</th>\n",
       "      <th>Expected</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>First_2_Dig</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>34</td>\n",
       "      <td>0.005697</td>\n",
       "      <td>0.004548</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>29</td>\n",
       "      <td>0.004859</td>\n",
       "      <td>0.004501</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>32</td>\n",
       "      <td>0.005362</td>\n",
       "      <td>0.004454</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>31</td>\n",
       "      <td>0.005194</td>\n",
       "      <td>0.004409</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>31</td>\n",
       "      <td>0.005194</td>\n",
       "      <td>0.004365</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Counts     Found  Expected\n",
       "First_2_Dig                            \n",
       "95               34  0.005697  0.004548\n",
       "96               29  0.004859  0.004501\n",
       "97               32  0.005362  0.004454\n",
       "98               31  0.005194  0.004409\n",
       "99               31  0.005194  0.004365"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f2d.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Assessing conformity"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### There are some conformity tests to evaluate if the data studied is a good fit to Benford's Law. Let us not confuse conformity tests with the digits tests themselves. The latter are the First Digit Test, Second Digit Test, First Two Digits Tests, and so on, ie, the result of processing the data and finding the respective position digits, whilst the former are statistical tests applied to the results of the latter and compared to critical values to establish conformity or not with Benford's Law.\n",
    "#### These conformity tests cam be divided for didatic purposes in:\n",
    "- #### Tests which result in a single (scalar) statistic about the whole sample. Some examples are:\n",
    "    - #### the **Chi-Square** test;\n",
    "    - #### the **Kolmogorov-Smirnov** test;\n",
    "    - #### the Mean Absolute Deviation (**MAD**) test; and\n",
    "- #### Tests that render one statistic for each of the studied leading digits' proportions in relation to the expexted Benford's ones. That's the case of the **Z statistic** for the proportions. We will start with this one.\n",
    "\n",
    "#### There is one more way to divide such tests:\n",
    "- #### Those which depend on the confidence level and on the number of records in the sample:\n",
    "    - #### the **Kolmogorov-Smirnov** test; and\n",
    "    - #### the **Z statistic** for the proportions.\n",
    "- #### Those which do not:\n",
    "    - #### the **Chi-Square** test; and\n",
    "    - #### The Mean Absolute Deviation (**MAD**) test."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### As mentioned, the Z scores test needs a confidence level so as to set a threshold for the deviations to be deemed relevant (above it) or not (below it). In the digits functions, you can turn it on by setting the parameter `confidence`, which will tell the function which confidence level to consider after calculating the Z score for each proportion."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Initialized sequence with 5968 registries.\n",
      "\n",
      "Test performed on 5968 registries.\n",
      "Discarded 0 records < 10 after preparation.\n",
      "\n",
      "The entries with the significant positive deviations are:\n",
      "\n",
      "             Expected     Found   Z_score\n",
      "First_2_Dig                              \n",
      "67           0.006434  0.010389  3.740056\n",
      "13           0.032185  0.039879  3.331418\n",
      "10           0.041393  0.049933  3.279619\n",
      "66           0.006531  0.009886  3.137524\n",
      "82           0.005264  0.007540  2.340301\n",
      "72           0.005990  0.008210  2.138736\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x972 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# For a significance of 5% (p <= 0.05), a confidence of 95%\n",
    "f2d = bf.first_digits(sp.l_r, digs=2, decimals=8, confidence=95)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Some things happened:\n",
    "- #### It printed a DataFrame with the **significant positive deviations**, in descending order of the Z scores. So, for a confidence level of 95%, the data to be considered for further investigation would be the records with First Two Digits **67**, **13**, **10**, **66**, **82** and **72**, whose propostions displayed a Z score higher than 1.96 (95%) **and** were positive, ie, were higher than the expected proportions. The function can also return **all** the relevant deviations or just the **negative** ones, by setting the `high-Z` parameter to 'all' or 'neg', respectively (see below).\n",
    "- #### In the plot, it added upper and lower boundaries to the Benford Expected line based on the level of confidence. Accordingly, it changed the colors of the bars whose proportions fell lower or higher than the drawn boundaries, for better vizualization.\n",
    "\n",
    "#### The *confidence* parameter takes the following (discrete) values other than *None*: 80, 85, 90, 95, 99, 99.9, 99.99, 99.999, 99.9999 and 99.99999."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### When you set the confidence level, the resulting DataFrame gets one more column, with the Z scores (bellow)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Counts</th>\n",
       "      <th>Found</th>\n",
       "      <th>Expected</th>\n",
       "      <th>Z_score</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>First_2_Dig</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>62</td>\n",
       "      <td>0.010389</td>\n",
       "      <td>0.006434</td>\n",
       "      <td>3.740056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>238</td>\n",
       "      <td>0.039879</td>\n",
       "      <td>0.032185</td>\n",
       "      <td>3.331418</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>298</td>\n",
       "      <td>0.049933</td>\n",
       "      <td>0.041393</td>\n",
       "      <td>3.279619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>126</td>\n",
       "      <td>0.021113</td>\n",
       "      <td>0.028029</td>\n",
       "      <td>3.197832</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>59</td>\n",
       "      <td>0.009886</td>\n",
       "      <td>0.006531</td>\n",
       "      <td>3.137524</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Counts     Found  Expected   Z_score\n",
       "First_2_Dig                                      \n",
       "67               62  0.010389  0.006434  3.740056\n",
       "13              238  0.039879  0.032185  3.331418\n",
       "10              298  0.049933  0.041393  3.279619\n",
       "15              126  0.021113  0.028029  3.197832\n",
       "66               59  0.009886  0.006531  3.137524"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f2d.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### To get a feeling of a less compliant sample, let us try the SPY closing prices instead of its returns, now with a confidence level of 99%. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Initialized sequence with 6026 registries.\n",
      "\n",
      "Test performed on 6026 registries.\n",
      "Discarded 0 records < 10 after preparation.\n",
      "\n",
      "The entries with the significant positive deviations are:\n",
      "\n",
      "             Expected     Found    Z_score\n",
      "First_2_Dig                               \n",
      "11           0.037789  0.135247  39.641478\n",
      "13           0.032185  0.111185  34.710863\n",
      "12           0.034762  0.115665  34.250363\n",
      "14           0.029963  0.084965  25.006243\n",
      "46           0.009340  0.030534  17.037044\n",
      "20           0.021189  0.045636  13.132371\n",
      "10           0.041393  0.074842  13.003059\n",
      "45           0.009545  0.021739   9.668882\n",
      "44           0.009760  0.019250   7.428122\n",
      "21           0.020203  0.029705   5.196428\n",
      "92           0.004695  0.008795   4.561719\n",
      "90           0.004799  0.007634   3.090972\n",
      "91           0.004746  0.007468   2.979729\n",
      "94           0.004596  0.007136   2.819974\n",
      "89           0.004853  0.007302   2.643275\n"
     ]
    },
    {
     "data": {
      "image/png": "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

Download .txt

gitextract_ip3h_u68/

├── .github/
│   └── workflows/
│       ├── pylint.yml
│       └── python-package.yml
├── .gitignore
├── .pylintrc
├── .readthedocs.yml
├── CITATION.cff
├── Demo.ipynb
├── LICENSE.txt
├── MANIFEST.in
├── README-pypi.md
├── README.md
├── __init__.py
├── benford/
│   ├── __init__.py
│   ├── benford.py
│   ├── checks.py
│   ├── constants.py
│   ├── expected.py
│   ├── reports.py
│   ├── stats.py
│   ├── utils.py
│   └── viz.py
├── data/
│   └── SPY.csv
├── docs/
│   ├── Makefile
│   ├── build/
│   │   ├── doctrees/
│   │   │   ├── api.doctree
│   │   │   ├── benford.doctree
│   │   │   ├── environment.pickle
│   │   │   ├── index.doctree
│   │   │   └── modules.doctree
│   │   └── html/
│   │       ├── .buildinfo
│   │       ├── _modules/
│   │       │   ├── benford/
│   │       │   │   ├── benford.html
│   │       │   │   ├── expected.html
│   │       │   │   ├── stats.html
│   │       │   │   ├── utils.html
│   │       │   │   └── viz.html
│   │       │   └── index.html
│   │       ├── _sources/
│   │       │   ├── api.rst.txt
│   │       │   ├── index.rst.txt
│   │       │   └── modules.rst.txt
│   │       ├── api.html
│   │       ├── genindex.html
│   │       ├── index.html
│   │       ├── modules.html
│   │       ├── objects.inv
│   │       ├── py-modindex.html
│   │       ├── search.html
│   │       └── searchindex.js
│   ├── make.bat
│   ├── requirements.txt
│   └── source/
│       ├── api.rst
│       ├── conf.py
│       ├── index.rst
│       └── modules.rst
├── setup.cfg
├── setup.py
└── tests/
    ├── __init__.py
    ├── conftest.py
    ├── test_checks.py
    ├── test_expected.py
    ├── test_stats.py
    └── test_utils.py

Download .txt

SYMBOL INDEX (272 symbols across 12 files)

FILE: benford/benford.py
  class Base (line 21) | class Base(DataFrame):
    method __init__ (line 40) | def __init__(self, data, decimals, sign='all', sec_order=False):
  class Test (line 89) | class Test(DataFrame):
    method __init__ (line 115) | def __init__(self, base, digs, confidence, limit_N=None, sec_order=Fal...
    method update_confidence (line 148) | def update_confidence(self, new_conf, check=True):
    method critical_values (line 164) | def critical_values(self):
    method show_plot (line 174) | def show_plot(self, save_plot=None, save_plot_kwargs=None):
    method report (line 193) | def report(self, high_Z='pos', show_plot=True,
  class Summ (line 223) | class Summ(DataFrame):
    method __init__ (line 231) | def __init__(self, base, test):
    method show_plot (line 250) | def show_plot(self, save_plot=None, save_plot_kwargs=None):
    method report (line 267) | def report(self, high_diff=None, show_plot=True,
  class Mantissas (line 290) | class Mantissas:
    method __init__ (line 300) | def __init__(self, data, confidence=95, limit_N=None):
    method stats (line 310) | def stats(self):
    method update_confidence (line 322) | def update_confidence(self, new_conf, check=True):
    method report (line 338) | def report(self, show_plot=True, save_plot=None, save_plot_kwargs=None):
    method show_plot (line 360) | def show_plot(self, figsize=(12, 6), save_plot=None, save_plot_kwargs=...
    method arc_test (line 378) | def arc_test(self, grid=True, figsize=12,
  class Benford (line 408) | class Benford(object):
    method __init__ (line 455) | def __init__(self, data, decimals=2, sign='all', confidence=95,
    method update_confidence (line 496) | def update_confidence(self, new_conf, tests=None):
    method all_confidences (line 531) | def all_confidences(self):
    method mantissas (line 542) | def mantissas(self):
    method sec_order (line 552) | def sec_order(self):
    method summation (line 579) | def summation(self):
  class Source (line 590) | class Source(DataFrame):
    method __init__ (line 614) | def __init__(self, data, decimals=2, sign='all', sec_order=False,
    method mantissas (line 663) | def mantissas(self, report=True, show_plot=True, figsize=(15, 8),
    method first_digits (line 697) | def first_digits(self, digs, confidence=None, high_Z='pos',
    method second_digit (line 822) | def second_digit(self, confidence=None, high_Z='pos',
    method last_two_digits (line 936) | def last_two_digits(self, confidence=None, high_Z='pos',
    method summation (line 1045) | def summation(self, digs=2, top=20, show_plot=True, save_plot=None,
    method duplicates (line 1101) | def duplicates(self, top_Rep=20, inform=None):
  class Roll_mad (line 1140) | class Roll_mad(object):
    method __init__ (line 1161) | def __init__(self, data, test, window, decimals=2, sign='all'):
    method show_plot (line 1176) | def show_plot(self, figsize=(15, 8), save_plot=None, save_plot_kwargs=...
  class Roll_mse (line 1194) | class Roll_mse(object):
    method __init__ (line 1214) | def __init__(self, data, test, window, decimals=2, sign='all'):
    method show_plot (line 1228) | def show_plot(self, figsize=(15, 8), save_plot=None, save_plot_kwargs=...
  function first_digits (line 1246) | def first_digits(data, digs, decimals=2, sign='all', verbose=True,
  function second_digit (line 1326) | def second_digit(data, decimals=2, sign='all', verbose=True,
  function last_two_digits (line 1403) | def last_two_digits(data, decimals=2, sign='all', verbose=True,
  function mantissas (line 1481) | def mantissas(data, report=True, show_plot=True, arc_test=True,
  function summation (line 1518) | def summation(data, digs=2, decimals=2, sign='all', top=20, verbose=True,
  function mad (line 1561) | def mad(data, test, decimals=2, sign='all', verbose=False):
  function mse (line 1592) | def mse(data, test, decimals=2, sign='all', verbose=False):
  function bhattacharyya_distance (line 1623) | def bhattacharyya_distance(data, test, decimals, sign="all", verbose=Fal...
  function kullback_leibler_divergence (line 1655) | def kullback_leibler_divergence(data, test, decimals, sign="all",
  function mad_summ (line 1688) | def mad_summ(data, test, decimals=2, sign='all', verbose=False):
  function rolling_mad (line 1721) | def rolling_mad(data, test, window, decimals=2, sign='all',
  function rolling_mse (line 1760) | def rolling_mse(data, test, window, decimals=2, sign='all',
  function duplicates (line 1799) | def duplicates(data, top_Rep=20, verbose=True, inform=None):
  function second_order (line 1843) | def second_order(data, test, decimals=2, sign='all', verbose=True, MAD=F...

FILE: benford/checks.py
  function _check_digs_ (line 6) | def _check_digs_(digs):
  function _check_test_ (line 15) | def _check_test_(test):
  function _check_decimals_ (line 36) | def _check_decimals_(decimals):
  function _check_sign_ (line 49) | def _check_sign_(sign):
  function _check_confidence_ (line 57) | def _check_confidence_(confidence):
  function _check_high_Z_ (line 65) | def _check_high_Z_(high_Z):
  function _check_num_array_ (line 74) | def _check_num_array_(data):

FILE: benford/expected.py
  class First (line 7) | class First(DataFrame):
    method __init__ (line 28) | def __init__(self, digs, plot=True, save_plot=None, save_plot_kwargs=N...
  class Second (line 41) | class Second(DataFrame):
    method __init__ (line 58) | def __init__(self, plot=True, save_plot=None, save_plot_kwargs=None):
  class LastTwo (line 70) | class LastTwo(DataFrame):
    method __init__ (line 87) | def __init__(self, num=False, plot=True, save_plot=None, save_plot_kwa...
  function _get_expected_digits_ (line 97) | def _get_expected_digits_(digs):
  function _gen_last_two_digits_ (line 114) | def _gen_last_two_digits_(num=False):
  function _gen_first_digits_ (line 137) | def _gen_first_digits_(digs):
  function _gen_second_digits_ (line 153) | def _gen_second_digits_():

FILE: benford/reports.py
  function _inform_ (line 5) | def _inform_(df, high_Z, conf):
  function _report_mad_ (line 44) | def _report_mad_(digs, MAD):
  function _report_KS_ (line 64) | def _report_KS_(KS, crit_KS):
  function _report_chi2_ (line 73) | def _report_chi2_(chi2, CRIT_CHI2):
  function _report_Z_ (line 82) | def _report_Z_(df, high_Z, crit_Z):
  function _report_summ_ (line 90) | def _report_summ_(test, high_diff):
  function _report_bhattac_coeff_ (line 103) | def _report_bhattac_coeff_(bhattac_coeff):
  function _report_bhattac_dist_ (line 109) | def _report_bhattac_dist_(bhattac_dist):
  function _report_kl_diverg_ (line 115) | def _report_kl_diverg_(kl_diverg):
  function _report_test_ (line 121) | def _report_test_(test, high=None, crit_vals=None):
  function _report_mantissa_ (line 146) | def _report_mantissa_(stats, confidence):
  function _deprecate_inform_ (line 167) | def _deprecate_inform_(verbose, inform):

FILE: benford/stats.py
  function Z_score (line 5) | def Z_score(frame, N):
  function chi_sq (line 20) | def chi_sq(frame, ddf, confidence, verbose=True):
  function chi_sq_2 (line 51) | def chi_sq_2(frame):
  function kolmogorov_smirnov (line 65) | def kolmogorov_smirnov(frame, confidence, N, verbose=True):
  function kolmogorov_smirnov_2 (line 99) | def kolmogorov_smirnov_2(frame):
  function _two_dist_ks_ (line 115) | def _two_dist_ks_(dist1, dist2, cummulative=True):
  function _mantissas_ks_ (line 134) | def _mantissas_ks_(mant_dist, confidence, sample_size):
  function mad (line 155) | def mad(frame, test, verbose=True):
  function mse (line 185) | def mse(frame, verbose=True):
  function _bhattacharyya_coefficient (line 203) | def _bhattacharyya_coefficient(dist_1, dist_2):
  function _bhattacharyya_distance_ (line 219) | def _bhattacharyya_distance_(dist_1, dist_2):
  function _kullback_leibler_divergence_ (line 237) | def _kullback_leibler_divergence_(dist_1, dist_2):

FILE: benford/utils.py
  function _set_N_ (line 9) | def _set_N_(len_df, limit_N):
  function get_mantissas (line 22) | def get_mantissas(arr):
  function input_data (line 35) | def input_data(given):
  function set_sign (line 63) | def set_sign(data, sign="all"):
  function get_times_10_power (line 78) | def get_times_10_power(data, decimals=2):
  function get_digs (line 97) | def get_digs(data, decimals=2, sign="all"):
  function get_found_proportions (line 126) | def get_found_proportions(data):
  function join_expect_found_diff (line 136) | def join_expect_found_diff(data, digs):
  function prepare (line 146) | def prepare(data, digs, limit_N=None, simple=False):
  function subtract_sorted (line 162) | def subtract_sorted(data):
  function prep_to_roll (line 171) | def prep_to_roll(start, test):
  function mad_to_roll (line 204) | def mad_to_roll(arr, Exp, ind):
  function mse_to_roll (line 215) | def mse_to_roll(arr, Exp, ind):

FILE: benford/viz.py
  function plot_expected (line 7) | def plot_expected(df, digs, save_plot=None, save_plot_kwargs=None):
  function _get_plot_args (line 47) | def _get_plot_args(digs):
  function plot_digs (line 66) | def plot_digs(df, x, y_Exp, y_Found, N, figsize, conf_Z, text_x=False,
  function plot_sum (line 137) | def plot_sum(df, figsize, li, text_x=False, save_plot=None, save_plot_kw...
  function plot_ordered_mantissas (line 179) | def plot_ordered_mantissas(col, figsize=(12, 12),
  function plot_mantissa_arc_test (line 217) | def plot_mantissa_arc_test(df, gravity_center, grid=True, figsize=12,
  function plot_roll_mse (line 264) | def plot_roll_mse(roll_series, figsize, save_plot=None, save_plot_kwargs...
  function plot_roll_mad (line 288) | def plot_roll_mad(roll_mad, figsize, save_plot=None, save_plot_kwargs=No...

FILE: tests/conftest.py
  function gen_N (line 12) | def gen_N():
  function gen_decimals (line 17) | def gen_decimals():
  function gen_N_lower (line 22) | def gen_N_lower(gen_N):
  function gen_array (line 27) | def gen_array(gen_N):
  function choose_digs_rand (line 33) | def choose_digs_rand():
  function choose_test (line 38) | def choose_test():
  function choose_confidence (line 43) | def choose_confidence():
  function gen_series (line 48) | def gen_series(gen_array):
  function gen_data_frame (line 53) | def gen_data_frame(gen_array):
  function gen_int_df (line 58) | def gen_int_df(gen_data_frame):
  function get_small_arrays (line 71) | def get_small_arrays(request):
  function small_str_foo_array (line 76) | def small_str_foo_array():
  function small_str_foo_series (line 81) | def small_str_foo_series():
  function gen_get_digs_df (line 86) | def gen_get_digs_df(gen_series, gen_decimals):
  function gen_proportions_F1D (line 91) | def gen_proportions_F1D(gen_get_digs_df):
  function gen_proportions_F2D (line 96) | def gen_proportions_F2D(gen_get_digs_df):
  function gen_proportions_F3D (line 101) | def gen_proportions_F3D(gen_get_digs_df):
  function gen_proportions_SD (line 106) | def gen_proportions_SD(gen_get_digs_df):
  function gen_proportions_L2D (line 111) | def gen_proportions_L2D(gen_get_digs_df):
  function gen_proportions_random_test (line 116) | def gen_proportions_random_test(choose_test, gen_get_digs_df):
  function gen_join_expect_found_diff_random_test (line 122) | def gen_join_expect_found_diff_random_test(gen_proportions_random_test):
  function gen_join_expect_found_diff_F1D (line 128) | def gen_join_expect_found_diff_F1D(gen_proportions_F1D):
  function gen_join_expect_found_diff_F2D (line 133) | def gen_join_expect_found_diff_F2D(gen_proportions_F2D):
  function gen_join_expect_found_diff_F3D (line 138) | def gen_join_expect_found_diff_F3D(gen_proportions_F3D):
  function gen_join_expect_found_diff_SD (line 143) | def gen_join_expect_found_diff_SD(gen_proportions_SD):
  function gen_join_expect_found_diff_L2D (line 148) | def gen_join_expect_found_diff_L2D(gen_proportions_L2D):
  function gen_linspaced_zero_one (line 153) | def gen_linspaced_zero_one(cuts:int=1000):
  function gen_mantissas_ks_dists (line 158) | def gen_mantissas_ks_dists(gen_array):
  function gen_mantissa_distribution (line 164) | def gen_mantissa_distribution():
  function get_mant_ks_types (line 179) | def get_mant_ks_types(request):
  function get_mant_ks_s (line 191) | def get_mant_ks_s(request):
  function gen_mantissa_distribution_len (line 196) | def gen_mantissa_distribution_len():
  function get_mant_ks_confs_limit_N (line 205) | def get_mant_ks_confs_limit_N(request):
  function gen_random_digs_and_proportions (line 213) | def gen_random_digs_and_proportions(gen_linspaced_zero_one, choose_digs_...

FILE: tests/test_checks.py
  class TestCheckDigs (line 8) | class TestCheckDigs():
    method test_digs_raise_msg (line 16) | def test_digs_raise_msg(self, dig, expectation):
    method test_check_digs_raise (line 23) | def test_check_digs_raise(self, dig, expectation):
    method test_check_digs_no_raise (line 31) | def test_check_digs_no_raise(self, dig, expectation):
  class TestCheckTest (line 36) | class TestCheckTest():
    method test_choose (line 42) | def test_choose(self, dig, expected):
    method test_raise (line 51) | def test_raise(self, dig, expectation):
    method test_None (line 55) | def test_None(self):
  class TestCheckDecimals (line 60) | class TestCheckDecimals():
    method test_positive_int (line 65) | def test_positive_int(self, pos_int, expected):
    method test_dec_raises (line 73) | def test_dec_raises(self, dec, expectation):
    method test_negative_int_msg (line 77) | def test_negative_int_msg(self):
    method test_infer (line 83) | def test_infer(self):
    method test_None_type (line 86) | def test_None_type(self):
  class TestCheckConfidence (line 91) | class TestCheckConfidence():
    method test_conf_raises (line 100) | def test_conf_raises(self, conf, expectation):
    method test_all_confidences (line 107) | def test_all_confidences(self, conf, expected):
  class TestCheckHighZ (line 111) | class TestCheckHighZ():
    method test_high_Z_raises (line 121) | def test_high_Z_raises(self, high_Z, expectation):
    method test_high_zs (line 129) | def test_high_zs(self, z, expected):
  class TestCheckNunmArray (line 133) | class TestCheckNunmArray():
    method test_arrays_to_float (line 143) | def test_arrays_to_float(self, arr):
    method test_small_arrays (line 146) | def test_small_arrays(self, get_small_arrays):
    method test_np_array_str (line 150) | def test_np_array_str(self, small_str_foo_array):
    method test_num_array_raises (line 163) | def test_num_array_raises(self, num_array, expectation):

FILE: tests/test_expected.py
  class TestGetExpectedDigits (line 5) | class TestGetExpectedDigits():
    method test_expected_types (line 12) | def test_expected_types(self, dig, expec_type):
    method test_expected_lenghts (line 20) | def test_expected_lenghts(self, dig, exp_len):
  class TestGenLastTwoDigits (line 24) | class TestGenLastTwoDigits():
    method test_types (line 29) | def test_types(self, num, arr_type):
  class TestGenDigits (line 34) | class TestGenDigits():
    method test_probs_sum_near_one (line 43) | def test_probs_sum_near_one(self, func, dig):
    method test_no_negative_prob (line 48) | def test_no_negative_prob(self, func, dig):
    method test_digs_sums (line 59) | def test_digs_sums(self, func, dig, exp_sum):
    method test_lengths (line 70) | def test_lengths(self, func, dig, exp_len):

FILE: tests/test_stats.py
  function test_Z_score_F1D (line 7) | def test_Z_score_F1D():
  class TestChiSquare (line 10) | class TestChiSquare():
    method test_conf_None (line 12) | def test_conf_None(self, gen_join_expect_found_diff_F1D, capsys):
    method test_random_conf_F1D (line 19) | def test_random_conf_F1D(self, gen_join_expect_found_diff_F1D,
    method test_random_conf_F2D (line 30) | def test_random_conf_F2D(self, gen_join_expect_found_diff_F2D,
    method test_random_conf_F3D (line 40) | def test_random_conf_F3D(self, gen_join_expect_found_diff_F3D,
    method test_random_conf_SD (line 50) | def test_random_conf_SD(self, gen_join_expect_found_diff_SD,
    method test_random_conf_L2D (line 60) | def test_random_conf_L2D(self, gen_join_expect_found_diff_L2D,
    method test_rand_test_rand_conf_verbose (line 70) | def test_rand_test_rand_conf_verbose(self, choose_confidence,
    method test_rand_test_conf_80 (line 80) | def test_rand_test_conf_80(self, gen_join_expect_found_diff_random_test):
    method test_rand_test_conf_85 (line 88) | def test_rand_test_conf_85(self, gen_join_expect_found_diff_random_test):
    method test_rand_test_conf_90 (line 96) | def test_rand_test_conf_90(self, gen_join_expect_found_diff_random_test):
    method test_rand_test_conf_95 (line 104) | def test_rand_test_conf_95(self, gen_join_expect_found_diff_random_test,
    method test_rand_test_conf_99 (line 116) | def test_rand_test_conf_99(self, gen_join_expect_found_diff_random_test):
    method test_rand_test_conf_999 (line 124) | def test_rand_test_conf_999(self, gen_join_expect_found_diff_random_te...
    method test_rand_test_conf_9999 (line 132) | def test_rand_test_conf_9999(self, gen_join_expect_found_diff_random_t...
    method test_rand_test_conf_99999 (line 140) | def test_rand_test_conf_99999(self, gen_join_expect_found_diff_random_...
    method test_rand_test_conf_999999 (line 148) | def test_rand_test_conf_999999(self, gen_join_expect_found_diff_random...
    method test_rand_test_conf_9999999 (line 156) | def test_rand_test_conf_9999999(self, gen_join_expect_found_diff_rando...
  class TestBhattacharyya (line 164) | class TestBhattacharyya():
    method test_coeff (line 166) | def test_coeff(self, gen_random_digs_and_proportions):
    method test_distance (line 173) | def test_distance(self, gen_random_digs_and_proportions):
  class TestKLDivergence (line 180) | class TestKLDivergence():
    method test_kld (line 182) | def test_kld(self, gen_random_digs_and_proportions):
  class TestTwoDistKS (line 189) | class TestTwoDistKS():
    method test_type (line 191) | def test_type(self, get_mant_ks_types):
    method test_more_equal_zero (line 196) | def test_more_equal_zero(self, get_mant_ks_s):
  class TestMantissasKS (line 201) | class TestMantissasKS:
    method test_confidence_limit_N (line 203) | def test_confidence_limit_N(self, get_mant_ks_confs_limit_N):

FILE: tests/test_utils.py
  class Test_set_N_ (line 6) | class Test_set_N_():
    method test_Limit_None (line 8) | def test_Limit_None(self, gen_N):
    method test_Limit_greater (line 11) | def test_Limit_greater(self, gen_N, gen_N_lower):
    method test_negative (line 14) | def test_negative(self, ):
    method test_float (line 18) | def test_float(self, ):
    method test_zero (line 22) | def test_zero(self, gen_N):
  class Test_get_mantissas (line 27) | class Test_get_mantissas():
    method test_less_than_1 (line 29) | def test_less_than_1(self, gen_array):
    method test_less_than_0 (line 32) | def test_less_than_0(self, gen_array):
  class Test_input_data (line 36) | class Test_input_data():
    method test_Series (line 38) | def test_Series(self, gen_series):
    method test_array (line 42) | def test_array(self, gen_array):
    method test_wrong_tuple (line 47) | def test_wrong_tuple(self, gen_array, gen_series, gen_data_frame):
    method test_df (line 53) | def test_df(self, gen_data_frame):
    method test_wrong_input_type (line 58) | def test_wrong_input_type(self, gen_array):
  class Test_set_sign (line 63) | class Test_set_sign():
    method test_all (line 65) | def test_all(self, gen_data_frame):
    method test_pos (line 69) | def test_pos(self, gen_data_frame):
    method test_neg (line 73) | def test_neg(self, gen_data_frame):
  class Test_get_times_10_power (line 78) | class Test_get_times_10_power():
    method test_2 (line 80) | def test_2(self, gen_data_frame):
    method test_8 (line 84) | def test_8(self, gen_data_frame):
    method test_0 (line 89) | def test_0(self, gen_int_df):
    method test_infer (line 94) | def test_infer(self, gen_data_frame):
  class Test_get_digs (line 100) | class Test_get_digs():
    method test_dec_8 (line 102) | def test_dec_8(self, gen_array):
    method test_dec_0 (line 109) | def test_dec_0(self, gen_array):
    method test_dec_2 (line 116) | def test_dec_2(self, gen_array):
    method test_dec_infer (line 123) | def test_dec_infer(self, gen_array):
  class Test_get_found_proportions (line 130) | class Test_get_found_proportions():
    method test_F1D (line 132) | def test_F1D(self, gen_proportions_F1D):
    method test_F2D (line 139) | def test_F2D(self, gen_proportions_F2D):
    method test_F3D (line 146) | def test_F3D(self, gen_proportions_F3D):
    method test_SD (line 153) | def test_SD(self, gen_proportions_SD):
    method test_L2D (line 160) | def test_L2D(self, gen_proportions_L2D):
  class Test_join_exp_found_diff (line 168) | class Test_join_exp_found_diff():
    method test_F1D (line 170) | def test_F1D(self, gen_proportions_F1D):
    method test_F2D (line 177) | def test_F2D(self, gen_proportions_F2D):
    method test_F3D (line 184) | def test_F3D(self, gen_proportions_F3D):
    method test_SD (line 191) | def test_SD(self, gen_proportions_SD):
    method test_L2D (line 198) | def test_L2D(self, gen_proportions_L2D):
  class Test_prepare (line 206) | class Test_prepare():
    method test_F1D_simple (line 208) | def test_F1D_simple(self, gen_series):
    method test_F2D_simple (line 212) | def test_F2D_simple(self, gen_series):
    method test_F3D_simple (line 216) | def test_F3D_simple(self, gen_series):
    method test_SD_simple (line 220) | def test_SD_simple(self, gen_series):
    method test_L2D_simple (line 224) | def test_L2D_simple(self, gen_series):
    method test_F1D (line 228) | def test_F1D(self, gen_series):
    method test_F2D (line 235) | def test_F2D(self, gen_series):
    method test_F3D (line 242) | def test_F3D(self, gen_series):
    method test_SD (line 249) | def test_SD(self, gen_series):
    method test_L2D (line 256) | def test_L2D(self, gen_series):
    method test_F1D_N (line 263) | def test_F1D_N(self, gen_N, gen_series):
    method test_F2D_N (line 270) | def test_F2D_N(self, gen_N, gen_series):
    method test_F3D_N (line 277) | def test_F3D_N(self, gen_N, gen_series):
    method test_SD_N (line 284) | def test_SD_N(self, gen_N, gen_series):
    method test_L2D_N (line 291) | def test_L2D_N(self, gen_N, gen_series):
  function test_subtract_sorted (line 299) | def test_subtract_sorted(gen_series):

Download .json

Condensed preview — 60 files, each showing path, character count, and a content snippet. Download the .json file or copy for the full structured content (2,393K chars).

[
  {
    "path": ".github/workflows/pylint.yml",
    "chars": 558,
    "preview": "name: Pylint\n\non: [push]\n\njobs:\n  build:\n\n    runs-on: ubuntu-latest\n\n    steps:\n      - uses: actions/checkout@v2\n     "
  },
  {
    "path": ".github/workflows/python-package.yml",
    "chars": 1298,
    "preview": "# This workflow will install Python dependencies, run tests and lint with a variety of Python versions\n# For more inform"
  },
  {
    "path": ".gitignore",
    "chars": 435,
    "preview": "# Compiled python modules.\n*.pyc\n\n__pycache__/\n\n# ipython notebook checkpoints\n*.ipynb_checkpoints\n\n# text editor backup"
  },
  {
    "path": ".pylintrc",
    "chars": 111,
    "preview": "[MASTER]\ndisable=\n    F0001, # No module named XXXXXXXX\n\n\n\nignored-classes=SQLObject,Registrant,scoped_session\n"
  },
  {
    "path": ".readthedocs.yml",
    "chars": 575,
    "preview": "# .readthedocs.yml\n# Read the Docs configuration file\n# See https://docs.readthedocs.io/en/stable/config-file/v2.html fo"
  },
  {
    "path": "CITATION.cff",
    "chars": 303,
    "preview": "cff-version: 1.2.0\nmessage: \"If you use this software, please cite it as below.\"\nauthors:\n- family-names: \"Milcent\"\n  gi"
  },
  {
    "path": "Demo.ipynb",
    "chars": 1075556,
    "preview": "{\n \"cells\": [\n  {\n   \"cell_type\": \"markdown\",\n   \"metadata\": {},\n   \"source\": [\n    \"# Benford  for Python\"\n   ]\n  },\n  "
  },
  {
    "path": "LICENSE.txt",
    "chars": 1498,
    "preview": "BSD 3-Clause License\n\nCopyright (c) 2014-2021, Marcel Milcent.\n\nRedistribution and use in source and binary forms, with "
  },
  {
    "path": "MANIFEST.in",
    "chars": 60,
    "preview": "include README.md\ninclude README-pypi.md\ninclude LICENSE.txt"
  },
  {
    "path": "README-pypi.md",
    "chars": 6722,
    "preview": "[![Downloads](https://pepy.tech/badge/benford-py)](https://pepy.tech/project/benford-py)\n\n# Benford for Python\n\n--------"
  },
  {
    "path": "README.md",
    "chars": 7315,
    "preview": "[![Downloads](https://pepy.tech/badge/benford-py)](https://pepy.tech/project/benford-py)\n\n# Benford for Python\n\n--------"
  },
  {
    "path": "__init__.py",
    "chars": 48,
    "preview": "'''Benfords law module'''\n__version__ = \"0.5.0\"\n"
  },
  {
    "path": "benford/__init__.py",
    "chars": 294,
    "preview": "\"\"\"\nBenford_py for Python is a module for application of Benford's Law\nto a sequence of numbers.\n\nDependent on pandas, n"
  },
  {
    "path": "benford/benford.py",
    "chars": 91587,
    "preview": "import warnings\nfrom pandas import Series, DataFrame\nfrom numpy import arange, log10, ones, abs, cos, sin, pi, mean\nfrom"
  },
  {
    "path": "benford/checks.py",
    "chars": 3028,
    "preview": "from pandas import Series\nfrom numpy import array, ndarray\nfrom .constants import DIGS, REV_DIGS, CONFS\n\n\ndef _check_dig"
  },
  {
    "path": "benford/constants.py",
    "chars": 3408,
    "preview": "DIGS = {1: 'F1D', 2: 'F2D', 3: 'F3D', 22: 'SD', -2: 'L2D'}\n\nSEC_ORDER_DIGS = {key: f'{val}_sec' for key, val in DIGS.ite"
  },
  {
    "path": "benford/expected.py",
    "chars": 6290,
    "preview": "from pandas import DataFrame\nfrom numpy import array, arange, log10\nfrom .checks import _check_digs_\nfrom .viz import pl"
  },
  {
    "path": "benford/reports.py",
    "chars": 6498,
    "preview": "import warnings\nfrom .constants import MAD_CONFORM\n\n\ndef _inform_(df, high_Z, conf):\n    \"\"\"Selects and sorts by the Z_s"
  },
  {
    "path": "benford/stats.py",
    "chars": 9108,
    "preview": "from numpy import abs as nabs, errstate, linspace, log, sqrt, where\nfrom .constants import CRIT_CHI2, CRIT_KS, MAD_CONFO"
  },
  {
    "path": "benford/utils.py",
    "chars": 6539,
    "preview": "from pandas import Series, DataFrame\nfrom numpy import array, arange, log10, ndarray\nfrom .expected import _get_expected"
  },
  {
    "path": "benford/viz.py",
    "chars": 12147,
    "preview": "from numpy import array, arange, maximum, sqrt, ones\nimport matplotlib.pyplot as plt\nfrom matplotlib.text import Annotat"
  },
  {
    "path": "data/SPY.csv",
    "chars": 419486,
    "preview": "Date,Open,High,Low,Close,Volume,Adj Close\n1993-01-29,43.9687,43.9687,43.75,43.9375,1003200,28.000838\n1993-02-01,43.9687,"
  },
  {
    "path": "docs/Makefile",
    "chars": 638,
    "preview": "# Minimal makefile for Sphinx documentation\n#\n\n# You can set these variables from the command line, and also\n# from the "
  },
  {
    "path": "docs/build/html/.buildinfo",
    "chars": 230,
    "preview": "# Sphinx build info version 1\n# This file hashes the configuration used when building these files. When it is not found,"
  },
  {
    "path": "docs/build/html/_modules/benford/benford.html",
    "chars": 285428,
    "preview": "\n\n<!DOCTYPE html>\n<html class=\"writer-html5\" lang=\"en\" >\n<head>\n  <meta charset=\"utf-8\" />\n  \n  <meta name=\"viewport\" co"
  },
  {
    "path": "docs/build/html/_modules/benford/expected.html",
    "chars": 23731,
    "preview": "\n\n<!DOCTYPE html>\n<html class=\"writer-html5\" lang=\"en\" >\n<head>\n  <meta charset=\"utf-8\" />\n  \n  <meta name=\"viewport\" co"
  },
  {
    "path": "docs/build/html/_modules/benford/stats.html",
    "chars": 33298,
    "preview": "\n\n<!DOCTYPE html>\n<html class=\"writer-html5\" lang=\"en\" >\n<head>\n  <meta charset=\"utf-8\" />\n  \n  <meta name=\"viewport\" co"
  },
  {
    "path": "docs/build/html/_modules/benford/utils.html",
    "chars": 28214,
    "preview": "\n\n<!DOCTYPE html>\n<!--[if IE 8]><html class=\"no-js lt-ie9\" lang=\"en\" > <![endif]-->\n<!--[if gt IE 8]><!--> <html class=\""
  },
  {
    "path": "docs/build/html/_modules/benford/viz.html",
    "chars": 60165,
    "preview": "\n\n<!DOCTYPE html>\n<html class=\"writer-html5\" lang=\"en\" >\n<head>\n  <meta charset=\"utf-8\" />\n  \n  <meta name=\"viewport\" co"
  },
  {
    "path": "docs/build/html/_modules/index.html",
    "chars": 4052,
    "preview": "\n\n<!DOCTYPE html>\n<html class=\"writer-html5\" lang=\"en\" >\n<head>\n  <meta charset=\"utf-8\" />\n  \n  <meta name=\"viewport\" co"
  },
  {
    "path": "docs/build/html/_sources/api.rst.txt",
    "chars": 557,
    "preview": "benford package\n===============\n\nbenford.benford module\n----------------------\n\n.. automodule:: benford.benford\n   :memb"
  },
  {
    "path": "docs/build/html/_sources/index.rst.txt",
    "chars": 537,
    "preview": "Welcome to benford_py's documentation!\n======================================\n\n.. toctree::\n   :maxdepth: 3\n   :caption:"
  },
  {
    "path": "docs/build/html/_sources/modules.rst.txt",
    "chars": 54,
    "preview": "benford\n=======\n\n.. toctree::\n   :maxdepth: 2\n\n   api\n"
  },
  {
    "path": "docs/build/html/api.html",
    "chars": 154918,
    "preview": "\n\n<!DOCTYPE html>\n<html class=\"writer-html5\" lang=\"en\" >\n<head>\n  <meta charset=\"utf-8\" />\n  \n  <meta name=\"viewport\" co"
  },
  {
    "path": "docs/build/html/genindex.html",
    "chars": 18029,
    "preview": "\n\n<!DOCTYPE html>\n<html class=\"writer-html5\" lang=\"en\" >\n<head>\n  <meta charset=\"utf-8\" />\n  \n  <meta name=\"viewport\" co"
  },
  {
    "path": "docs/build/html/index.html",
    "chars": 6358,
    "preview": "\n\n<!DOCTYPE html>\n<html class=\"writer-html5\" lang=\"en\" >\n<head>\n  <meta charset=\"utf-8\" />\n  \n  <meta name=\"viewport\" co"
  },
  {
    "path": "docs/build/html/modules.html",
    "chars": 5367,
    "preview": "\n\n<!DOCTYPE html>\n<html class=\"writer-html5\" lang=\"en\" >\n<head>\n  <meta charset=\"utf-8\" />\n  \n  <meta name=\"viewport\" co"
  },
  {
    "path": "docs/build/html/py-modindex.html",
    "chars": 5060,
    "preview": "\n\n<!DOCTYPE html>\n<html class=\"writer-html5\" lang=\"en\" >\n<head>\n  <meta charset=\"utf-8\" />\n  \n  <meta name=\"viewport\" co"
  },
  {
    "path": "docs/build/html/search.html",
    "chars": 4231,
    "preview": "\n\n<!DOCTYPE html>\n<html class=\"writer-html5\" lang=\"en\" >\n<head>\n  <meta charset=\"utf-8\" />\n  \n  <meta name=\"viewport\" co"
  },
  {
    "path": "docs/build/html/searchindex.js",
    "chars": 6824,
    "preview": "Search.setIndex({docnames:[\"api\",\"index\",\"modules\"],envversion:{\"sphinx.domains.c\":2,\"sphinx.domains.changeset\":1,\"sphin"
  },
  {
    "path": "docs/make.bat",
    "chars": 799,
    "preview": "@ECHO OFF\r\n\r\npushd %~dp0\r\n\r\nREM Command file for Sphinx documentation\r\n\r\nif \"%SPHINXBUILD%\" == \"\" (\r\n\tset SPHINXBUILD=sp"
  },
  {
    "path": "docs/requirements.txt",
    "chars": 23,
    "preview": "pandas\nnumpy\nmatplotlib"
  },
  {
    "path": "docs/source/api.rst",
    "chars": 557,
    "preview": "benford package\n===============\n\nbenford.benford module\n----------------------\n\n.. automodule:: benford.benford\n   :memb"
  },
  {
    "path": "docs/source/conf.py",
    "chars": 2088,
    "preview": "# Configuration file for the Sphinx documentation builder.\n#\n# This file only contains a selection of the most common op"
  },
  {
    "path": "docs/source/index.rst",
    "chars": 537,
    "preview": "Welcome to benford_py's documentation!\n======================================\n\n.. toctree::\n   :maxdepth: 3\n   :caption:"
  },
  {
    "path": "docs/source/modules.rst",
    "chars": 54,
    "preview": "benford\n=======\n\n.. toctree::\n   :maxdepth: 2\n\n   api\n"
  },
  {
    "path": "setup.cfg",
    "chars": 39,
    "preview": "[metadata]\ndescription-file = README.md"
  },
  {
    "path": "setup.py",
    "chars": 1454,
    "preview": "''' Setup for benford's module'''\nfrom os import path\nfrom setuptools import setup\n\nthis_directory = path.abspath(path.d"
  },
  {
    "path": "tests/__init__.py",
    "chars": 0,
    "preview": ""
  },
  {
    "path": "tests/conftest.py",
    "chars": 5555,
    "preview": "from random import choice\nimport pytest\nimport numpy as np\nimport pandas as pd\nfrom ..benford import utils as ut\nfrom .."
  },
  {
    "path": "tests/test_checks.py",
    "chars": 5549,
    "preview": "from contextlib import suppress as do_not_raise\nimport pytest\nfrom pytest import raises\nfrom ..benford import checks as "
  },
  {
    "path": "tests/test_expected.py",
    "chars": 2409,
    "preview": "import pytest\nfrom ..benford import expected as ex\n\n\nclass TestGetExpectedDigits():\n\n    expected_types = [\n        (x, "
  },
  {
    "path": "tests/test_stats.py",
    "chars": 8453,
    "preview": "import pytest\nfrom numpy import float_\nfrom ..benford import stats as st\nfrom ..benford.constants import CRIT_CHI2, CRIT"
  },
  {
    "path": "tests/test_utils.py",
    "chars": 10593,
    "preview": "import pytest\nimport pandas as pd\nfrom ..benford import utils as ut\n\n\nclass Test_set_N_():\n        \n    def test_Limit_N"
  }
]

// ... and 6 more files (download for full content)

About this extraction

This page contains the full source code of the milcent/benford_py GitHub repository, extracted and formatted as plain text for AI agents and large language models (LLMs). The extraction includes 60 files (2.2 MB), approximately 584.6k tokens, and a symbol index with 272 extracted functions, classes, methods, constants, and types. Use this with OpenClaw, Claude, ChatGPT, Cursor, Windsurf, or any other AI tool that accepts text input. You can copy the full output to your clipboard or download it as a .txt file.

Extracted by GitExtract — free GitHub repo to text converter for AI. Built by Nikandr Surkov.

Extract another repo