2020-08-03T14:45:23,491 Created temporary directory: /tmp/pip-ephem-wheel-cache-c2ro9x8m
2020-08-03T14:45:23,495 Created temporary directory: /tmp/pip-req-tracker-qi8cr40e
2020-08-03T14:45:23,496 Initialized build tracking at /tmp/pip-req-tracker-qi8cr40e
2020-08-03T14:45:23,496 Created build tracker: /tmp/pip-req-tracker-qi8cr40e
2020-08-03T14:45:23,497 Entered build tracker: /tmp/pip-req-tracker-qi8cr40e
2020-08-03T14:45:23,498 Created temporary directory: /tmp/pip-wheel-p7jn2a4e
2020-08-03T14:45:23,515 1 location(s) to search for versions of odus:
2020-08-03T14:45:23,515 * https://pypi.org/simple/odus/
2020-08-03T14:45:23,516 Fetching project page and analyzing links: https://pypi.org/simple/odus/
2020-08-03T14:45:23,516 Getting page https://pypi.org/simple/odus/
2020-08-03T14:45:23,519 Found index url https://pypi.org/simple
2020-08-03T14:45:23,725   Skipping link: No binaries permitted for odus: https://files.pythonhosted.org/packages/54/5c/314e6bd75cce4b45cfa5ff43a70ae97783a9f4c55b54b5874421db76e2c8/odus-0.0.1-py3-none-any.whl#sha256=0e3ab1c76df9b648112534048773c9611fbaa38831575c724ec6e59144fac86f (from https://pypi.org/simple/odus/)
2020-08-03T14:45:23,726   Found link https://files.pythonhosted.org/packages/64/96/703d2db924bce4c35daef3f2289bf3c4d793e1e1ed655969728a0ebfae80/odus-0.0.1.tar.gz#sha256=d8c345a55f5b07f1c53d62f7f72f5346df441c410f952eb7184f52367c4ae925 (from https://pypi.org/simple/odus/), version: 0.0.1
2020-08-03T14:45:23,727   Skipping link: No binaries permitted for odus: https://files.pythonhosted.org/packages/f6/ba/337ae3f8204b56312a4d55be33b6865da599a0a526b55d0ce394d9c9d8cb/odus-0.0.3-py3-none-any.whl#sha256=10d8a8a9aef138da0d54948a147d0ba0b410c9fca2bc98286c6a8d13c6e1ff47 (from https://pypi.org/simple/odus/)
2020-08-03T14:45:23,727   Found link https://files.pythonhosted.org/packages/4d/d9/c620386320ec5d657cf3865bbc4df315bdf0eeb1709789cddb96ae428371/odus-0.0.3.tar.gz#sha256=b02a21f2e2efa307cfa66b6ad89faa9029527664777ae4c43b25335fee0c5362 (from https://pypi.org/simple/odus/), version: 0.0.3
2020-08-03T14:45:23,728   Skipping link: No binaries permitted for odus: https://files.pythonhosted.org/packages/e4/99/517b778c38af9c1bfdebd76679a5435554b4f56f15a9a7951b9eec694b7d/odus-0.0.5-py3-none-any.whl#sha256=53222cf6d230a133df6d16a0e95de840c3ac70f7f32580e88ae7641638b87ac4 (from https://pypi.org/simple/odus/)
2020-08-03T14:45:23,728   Found link https://files.pythonhosted.org/packages/6a/84/a24a9f511406da5ebee3411751582d9a38ac77c57bb792b734ffe8710bc7/odus-0.0.5.tar.gz#sha256=761f997aa4eac4d7b120f2a55f46976e87568264a469be229ab8f556bb4b12b2 (from https://pypi.org/simple/odus/), version: 0.0.5
2020-08-03T14:45:23,729   Skipping link: No binaries permitted for odus: https://files.pythonhosted.org/packages/55/7e/5433d19e19618fc86242778ff09df3307c46a9f95f6e52f8ab69a22de31e/odus-0.0.6-py3-none-any.whl#sha256=0dc996d310b924f8055e79ede0d24f62eb0e24a16f4ad50d17f03a156ff736ab (from https://pypi.org/simple/odus/)
2020-08-03T14:45:23,729   Found link https://files.pythonhosted.org/packages/be/ea/a3d88b705ff73b7a7435335d1b2d6aa6202a78a9537d9ed219e4b77e428b/odus-0.0.6.tar.gz#sha256=e43b83b217218c592b938968f034b6f4045d5511baef54ec3c0b1bda60dbcad3 (from https://pypi.org/simple/odus/), version: 0.0.6
2020-08-03T14:45:23,738 Given no hashes to check 1 links for project 'odus': discarding no candidates
2020-08-03T14:45:23,739 Using version 0.0.6 (newest of versions: 0.0.6)
2020-08-03T14:45:23,744 Collecting odus==0.0.6
2020-08-03T14:45:23,748   Created temporary directory: /tmp/pip-unpack-nw9wyybq
2020-08-03T14:45:23,853   Downloading odus-0.0.6.tar.gz (15 kB)
2020-08-03T14:45:23,931   Added odus==0.0.6 from https://files.pythonhosted.org/packages/be/ea/a3d88b705ff73b7a7435335d1b2d6aa6202a78a9537d9ed219e4b77e428b/odus-0.0.6.tar.gz#sha256=e43b83b217218c592b938968f034b6f4045d5511baef54ec3c0b1bda60dbcad3 to build tracker '/tmp/pip-req-tracker-qi8cr40e'
2020-08-03T14:45:23,933     Running setup.py (path:/tmp/pip-wheel-p7jn2a4e/odus/setup.py) egg_info for package odus
2020-08-03T14:45:23,934     Created temporary directory: /tmp/pip-pip-egg-info-k3z809f5
2020-08-03T14:45:23,935     Running command python setup.py egg_info
2020-08-03T14:45:25,588     Setup params -------------------------------------------------------
2020-08-03T14:45:25,589     {
2020-08-03T14:45:25,589       "name": "odus",
2020-08-03T14:45:25,590       "version": "0.0.6",
2020-08-03T14:45:25,590       "url": "https://github.com/thorwhalen/odus",
2020-08-03T14:45:25,591       "author": "Thor Whalen",
2020-08-03T14:45:25,591       "author_email": "thorwhalen1@gmail.com",
2020-08-03T14:45:25,592       "license": "MIT",
2020-08-03T14:45:25,592       "include_package_data": true,
2020-08-03T14:45:25,593       "platforms": "any",
2020-08-03T14:45:25,593       "long_description": "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#Introduction\" data-toc-modified-id=\"Introduction-1\"><span class=\"toc-item-num\">1&nbsp;&nbsp;</span>Introduction</a></span></li><li><span><a href=\"#Installation\" data-toc-modified-id=\"Installation-2\"><span class=\"toc-item-num\">2&nbsp;&nbsp;</span>Installation</a></span></li><li><span><a href=\"#Getting-some-resources\" data-toc-modified-id=\"Getting-some-resources-3\"><span class=\"toc-item-num\">3&nbsp;&nbsp;</span>Getting some resources</a></span></li><li><span><a href=\"#Poking-around\" data-toc-modified-id=\"Poking-around-4\"><span class=\"toc-item-num\">4&nbsp;&nbsp;</span>Poking around</a></span><ul class=\"toc-item\"><li><span><a href=\"#df_store\" data-toc-modified-id=\"df_store-4.1\"><span class=\"toc-item-num\">4.1&nbsp;&nbsp;</span>df_store</a></span></li><li><span><a href=\"#Viewing-trajectories\" data-toc-modified-id=\"Viewing-trajectories-4.2\"><span class=\"toc-item-num\">4.2&nbsp;&nbsp;</span>Viewing trajectories</a></span><ul class=\"toc-item\"><li><span><a href=\"#Viewing-a-single-trajectory\" data-toc-modified-id=\"Viewing-a-single-trajectory-4.2.1\"><span class=\"toc-item-num\">4.2.1&nbsp;&nbsp;</span>Viewing a single trajectory</a></span></li><li><span><a href=\"#Flip-over-all-(or-some)-trajectories\" data-toc-modified-id=\"Flip-over-all-(or-some)-trajectories-4.2.2\"><span class=\"toc-item-num\">4.2.2&nbsp;&nbsp;</span>Flip over all (or some) trajectories</a></span></li></ul></li><li><span><a href=\"#Making-a-pdf-of-trajectories\" data-toc-modified-id=\"Making-a-pdf-of-trajectories-4.3\"><span class=\"toc-item-num\">4.3&nbsp;&nbsp;</span>Making a pdf of trajectories</a></span></li><li><span><a href=\"#Demo-s-and-v\" data-toc-modified-id=\"Demo-s-and-v-4.4\"><span class=\"toc-item-num\">4.4&nbsp;&nbsp;</span>Demo s and v</a></span></li><li><span><a href=\"#cstore\" data-toc-modified-id=\"cstore-4.5\"><span class=\"toc-item-num\">4.5&nbsp;&nbsp;</span>cstore</a></span></li><li><span><a href=\"#pstore\" data-toc-modified-id=\"pstore-4.6\"><span class=\"toc-item-num\">4.6&nbsp;&nbsp;</span>pstore</a></span></li><li><span><a href=\"#Scrap-place\" data-toc-modified-id=\"Scrap-place-4.7\"><span class=\"toc-item-num\">4.7&nbsp;&nbsp;</span>Scrap place</a></span></li></ul></li><li><span><a href=\"#Potential-Calculus-Experimentations\" data-toc-modified-id=\"Potential-Calculus-Experimentations-5\"><span class=\"toc-item-num\">5&nbsp;&nbsp;</span>Potential Calculus Experimentations</a></span></li><li><span><a href=\"#Extra-scrap\" data-toc-modified-id=\"Extra-scrap-6\"><span class=\"toc-item-num\">6&nbsp;&nbsp;</span>Extra scrap</a></span></li></ul></div>\n\n\n```python\n# %load_ext autoreload\n# %autoreload 2\n```\n\n# Introduction\n\nODUS (for Older Drug User Study) contains data and tools to study the drug use of older drug users.\n\nEssentially, there are these are tools:\n\n- To get prepared data on the 119 \"trajectories\" describing 31 variables (drug use, social, etc.) over time of 119 different respondents.\n\n- To vizualize these trajectories in various ways\n\n- To create pdfs of any selection of these trajectories and variables\n\n- To make count tables for any combinations of the variables: Essential step of any Markovian or Bayesian analysis.\n\n- To make probability (joint or conditional) tables from any combination of the variables\n\n- To operate on these count and probability tables, thus enabling inference operations\n\n\n# Installation\n\nYou need to have python 3.7+ to run this notebook.\n\nAnd you'll need to have `odus`, which you get by doing\n\n```\npip install odus\n```\n\n(And if you don't have pip then, well... how to put it... ha ha ha!)\n\nBut if you're the type, you can also just get the source from `https://github.com/thorwhalen/odus`. \n\nOh, and pull requests etc. are welcome!\n\nStars, likes, references, and coffee also welcome.\n\nAnd if you want to donate: Donate to a charity that will help the people understand and make policies surrounding the use of substances.\n\nA simple flowchart about the architecture:\n\n<img src=\"odus_store.png\" width=800 height=200>\n\n# Getting some resources\n\n\n```python\nfrom matplotlib.pylab import *\nfrom numpy import *\nimport seaborn as sns\n\nimport os\nfrom py2store.stores.local_store import RelativePathFormatStore\nfrom py2store.mixins import ReadOnlyMixin\nfrom py2store.base import Store\n\n\nfrom io import BytesIO\nfrom spyn.ppi.pot import Pot, ProbPot\nfrom collections import UserDict, Counter\nimport numpy as np\nimport pandas as pd\n\nfrom ut.ml.feature_extraction.sequential_var_sets import PVar, VarSet, DfData, VarSetFactory\nfrom IPython.display import Image\n\nfrom odus.analysis_utils import *\n\nfrom odus.dacc import DfStore, counts_of_kps, Dacc, VarSetCountsStore, \\\n    mk_pvar_struct, PotStore, _commun_columns_of_dfs, Struct, mk_pvar_str_struct, VarStr\n\nfrom odus.plot_utils import plot_life_course\n```\n\n\n```python\nfrom odus import data_dir, data_path_of\nsurvey_dir = data_dir\ndata_dir\n```\n\n\n\n\n    '/D/Dropbox/dev/p3/proj/odus/odus/data'\n\n\n\n\n```python\ndf_store = DfStore(data_dir + '/{}.xlsx')\nlen(df_store)\ncstore = VarSetCountsStore(df_store)\nv = mk_pvar_struct(df_store, only_for_cols_in_all_dfs=True)\ns = mk_pvar_str_struct(v)\nf, df = cstore.df_store.head()\npstore = PotStore(df_store)\n```\n\n# Poking around\n\n## df_store\n\nA df_store is a key-value store where the key is the xls file and the value is the prepared dataframe\n\n\n```python\nlen(df_store)\n```\n\n\n\n\n    119\n\n\n\n\n```python\nit = iter(df_store.values())\nfor i in range(5):  # skip five first\n    _ = next(it)\ndf = next(it)  # get the one I want\ndf.head(3)\n```\n\n\n\n\n<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>category</th>\n      <th>RURAL</th>\n      <th>SUBURBAN</th>\n      <th>URBAN/CITY</th>\n      <th>HOMELESS</th>\n      <th>INCARCERATION</th>\n      <th>WORK</th>\n      <th>SON/DAUGHTER</th>\n      <th>SIBLING</th>\n      <th>FATHER/MOTHER</th>\n      <th>SPOUSE</th>\n      <th>...</th>\n      <th>METHAMPHETAMINE</th>\n      <th>AS PRESCRIBED OPIOID</th>\n      <th>NOT AS PRESCRIBED OPIOID</th>\n      <th>HEROIN</th>\n      <th>OTHER OPIOID</th>\n      <th>INJECTED</th>\n      <th>IN TREATMENT</th>\n      <th>Selects States below</th>\n      <th>Georgia</th>\n      <th>Pennsylvania</th>\n    </tr>\n    <tr>\n      <th>age</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      <th></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      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>11</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>13</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n<p>3 rows \u00d7 31 columns</p>\n</div>\n\n\n\n\n```python\nprint(df.columns.values)\n```\n\n    ['RURAL' 'SUBURBAN' 'URBAN/CITY' 'HOMELESS' 'INCARCERATION' 'WORK'\n     'SON/DAUGHTER' 'SIBLING' 'FATHER/MOTHER' 'SPOUSE'\n     'OTHER (WHO?, FILL IN BRACKETS HERE)' 'FRIEND USER' 'FRIEND NON USER'\n     'MENTAL ILLNESS' 'PHYSICAL ILLNESS' 'LOSS OF LOVED ONE' 'TOBACCO'\n     'MARIJUANA' 'ALCOHOL' 'HAL/LSD/XTC/CLUBDRUG' 'COCAINE/CRACK'\n     'METHAMPHETAMINE' 'AS PRESCRIBED OPIOID' 'NOT AS PRESCRIBED OPIOID'\n     'HEROIN' 'OTHER OPIOID' 'INJECTED' 'IN TREATMENT' 'Selects States below'\n     'Georgia' 'Pennsylvania']\n\n\n\n```python\nt = df[['ALCOHOL', 'TOBACCO']]\nt.head(3)\n```\n\n\n\n\n<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>category</th>\n      <th>ALCOHOL</th>\n      <th>TOBACCO</th>\n    </tr>\n    <tr>\n      <th>age</th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>11</th>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>13</th>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n</div>\n\n\n\n\n```python\nc = Counter()\nfor i, r in t.iterrows():\n    c.update([tuple(r.to_list())])\nc\n```\n\n\n\n\n    Counter({(0, 0): 6, (1, 0): 4, (1, 1): 9, (0, 1): 2})\n\n\n\n\n```python\ndef count_tuples(dataframe):\n    c = Counter()\n    for i, r in dataframe.iterrows():\n        c.update([tuple(r.to_list())])\n    return c\n```\n\n\n```python\nfields = ['ALCOHOL', 'TOBACCO']\n# do it for every one\nc = Counter()\nfor df in df_store.values():\n    c.update(count_tuples(df[fields]))\nc\n```\n\n\n\n\n    Counter({(0, 1): 903, (1, 1): 1343, (0, 0): 240, (1, 0): 179})\n\n\n\n\n```python\npd.Series(c)\n```\n\n\n\n\n    0  1     903\n    1  1    1343\n    0  0     240\n    1  0     179\n    dtype: int64\n\n\n\n\n```python\n# Powerful! You can use that with several pairs and get some nice probabilities. Look up Naive Bayes.\n```\n\n## Viewing trajectories\n\n\n```python\nimport itertools\nfrom functools import partial\nfrom odus.util import write_images\nfrom odus.plot_utils import plot_life, life_plots, write_trajectories_to_file\n\nihead = lambda it: itertools.islice(it, 0, 5)\n```\n\n### Viewing a single trajectory\n\n\n```python\nk = next(iter(df_store))  # get the first key\nprint(f\"k: {k}\")  # print it\nplot_life(df_store[k])  # plot the trajectory\n```\n\n    k: surveys/B24.xlsx\n\n\n\n![png](img/output_26_1.png)\n\n\n\n```python\nplot_life(df_store[k], fields=[s.in_treatment, s.injected])  # only want two fields\n```\n\n\n![png](img/output_27_0.png)\n\n\n### Flip over all (or some) trajectories\n\n\n```python\ngen = life_plots(df_store)\n```\n\n\n```python\nnext(gen)  # launch to get the next trajectory\n```\n\n\n\n\n    <matplotlib.axes._subplots.AxesSubplot at 0x12b21f070>\n\n\n\n\n![png](img/output_30_1.png)\n\n\nGet three trajectories, but only over two fields.\n\n\n```python\n# fields = [s.in_treatment, s.injected]\nfields = [s.physical_illness, s.as_prescribed_opioid, s.heroin, s.other_opioid]\nkeys = list(df_store)[:10]\n# print(f\"keys={keys}\")\naxs = [x for x in life_plots(df_store, fields, keys=keys)];\n```\n\n\n![png](img/output_32_0.png)\n\n\n\n![png](img/output_32_1.png)\n\n\n\n![png](img/output_32_2.png)\n\n\n\n![png](img/output_32_3.png)\n\n\n\n![png](img/output_32_4.png)\n\n\n\n![png](img/output_32_5.png)\n\n\n\n![png](img/output_32_6.png)\n\n\n\n![png](img/output_32_7.png)\n\n\n\n![png](img/output_32_8.png)\n\n\n\n![png](img/output_32_9.png)\n\n\n## Making a pdf of trajectories\n\n\n```python\nwrite_trajectories_to_file(df_store, fields, keys, fp='three_respondents_two_fields.pdf');\n```\n\n\n```python\nwrite_trajectories_to_file(df_store, fp='all_respondents_all_fields.pdf');\n```\n\n\n```python\n \n```\n\n## Demo s and v\n\n\n```python\nprint(list(filter(lambda x: not x.startswith('__'), dir(s))))\n```\n\n    ['alcohol', 'as_prescribed_opioid', 'cocaine_crack', 'father_mother', 'hal_lsd_xtc_clubdrug', 'heroin', 'homeless', 'in_treatment', 'incarceration', 'injected', 'loss_of_loved_one', 'marijuana', 'mental_illness', 'methamphetamine', 'not_as_prescribed_opioid', 'other_opioid', 'physical_illness', 'rural', 'sibling', 'son_daughter', 'suburban', 'tobacco', 'urban_city', 'work']\n\n\n\n```python\ns.heroin\n```\n\n\n\n\n    'HEROIN'\n\n\n\n\n```python\nv.heroin\n```\n\n\n\n\n    PVar('HEROIN', 0)\n\n\n\n\n```python\nv.heroin - 1\n```\n\n\n\n\n    PVar('HEROIN', -1)\n\n\n\n## cstore\n\n\n```python\n# cstore[v.alcohol, v.tobacco]\ncstore[v.as_prescribed_opioid-1, v.heroin]\n```\n\n\n\n\n    Counter({(0, 0): 1026, (1, 0): 264, (0, 1): 1108, (1, 1): 148})\n\n\n\n\n```python\npd.Series(cstore[v.as_prescribed_opioid-1, v.heroin])\n```\n\n\n\n\n    0  0    1026\n    1  0     264\n    0  1    1108\n    1  1     148\n    dtype: int64\n\n\n\n\n```python\ncstore[v.alcohol, v.tobacco, v.heroin]\n```\n\n\n\n\n    Counter({(0, 0, 1): 427,\n             (1, 0, 1): 656,\n             (1, 1, 1): 687,\n             (0, 0, 0): 189,\n             (0, 1, 1): 476,\n             (0, 1, 0): 51,\n             (1, 0, 0): 133,\n             (1, 1, 0): 46})\n\n\n\n\n```python\ncstore[v.alcohol-1, v.alcohol]\n```\n\n\n\n\n    Counter({(0, 0): 994, (1, 1): 1375, (1, 0): 90, (0, 1): 87})\n\n\n\n\n```python\ncstore[v.alcohol-1, v.alcohol, v.tobacco]\n```\n\n\n\n\n    Counter({(0, 0, 1): 807,\n             (1, 1, 1): 1220,\n             (1, 0, 0): 26,\n             (0, 1, 1): 76,\n             (0, 0, 0): 187,\n             (1, 1, 0): 155,\n             (0, 1, 0): 11,\n             (1, 0, 1): 64})\n\n\n\n\n```python\nt = pd.Series(cstore[v.alcohol-1, v.alcohol, v.tobacco])\nt.loc[t.index]\n```\n\n\n\n\n    <pandas.core.indexing._LocIndexer at 0x130955db0>\n\n\n\n## pstore\n\n\n```python\nt = pstore[s.alcohol-1, s.alcohol]\nt\n```\n\n\n\n\n                       pval\n    ALCOHOL-1 ALCOHOL      \n    0         0         994\n              1          87\n    1         0          90\n              1        1375\n\n\n\n\n```python\nt.tb\n```\n\n\n\n\n<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>ALCOHOL-1</th>\n      <th>ALCOHOL</th>\n      <th>pval</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>0</td>\n      <td>994</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>1</td>\n      <td>87</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>0</td>\n      <td>90</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>1</td>\n      <td>1375</td>\n    </tr>\n  </tbody>\n</table>\n</div>\n\n\n\n\n```python\nt / []\n```\n\n\n\n\n                           pval\n    ALCOHOL-1 ALCOHOL          \n    0         0        0.390416\n              1        0.034171\n    1         0        0.035350\n              1        0.540063\n\n\n\n\n```python\nt[s.alcohol-1]\n```\n\n\n\n\n               pval\n    ALCOHOL-1      \n    0          1081\n    1          1465\n\n\n\n\n```python\nt / t[s.alcohol-1]  # cond prob!\n```\n\n\n\n\n                           pval\n    ALCOHOL-1 ALCOHOL          \n    0         0        0.919519\n              1        0.080481\n    1         0        0.061433\n              1        0.938567\n\n\n\n\n```python\ntt = pstore[s.alcohol, s.tobacco]\ntt\n```\n\n\n\n\n                     pval\n    ALCOHOL TOBACCO      \n    0       0         240\n            1         903\n    1       0         179\n            1        1343\n\n\n\n\n```python\ntt / tt[s.alcohol]\n```\n\n\n\n\n                         pval\n    ALCOHOL TOBACCO          \n    0       0        0.209974\n            1        0.790026\n    1       0        0.117608\n            1        0.882392\n\n\n\n\n```python\ntt / tt[s.tobacco]\n```\n\n\n\n\n                         pval\n    ALCOHOL TOBACCO          \n    0       0        0.572792\n    1       0        0.427208\n    0       1        0.402048\n    1       1        0.597952\n\n\n\n\n```python\n\n```\n\n## Scrap place\n\n\n```python\nt = pstore[s.as_prescribed_opioid-1, s.heroin-1, s.heroin]\nt\n\n```\n\n\n\n\n                                            pval\n    AS PRESCRIBED OPIOID-1 HEROIN-1 HEROIN      \n    0                      0        0        927\n                                    1        172\n                           1        0         99\n                                    1        936\n    1                      0        0        249\n                                    1         33\n                           1        0         15\n                                    1        115\n\n\n\n\n```python\ntt = t / t[s.as_prescribed_opioid-1, s.heroin-1]  # cond prob!\ntt\n```\n\n\n\n\n                                                pval\n    AS PRESCRIBED OPIOID-1 HEROIN-1 HEROIN          \n    0                      0        0       0.843494\n                                    1       0.156506\n                           1        0       0.095652\n                                    1       0.904348\n    1                      0        0       0.882979\n                                    1       0.117021\n                           1        0       0.115385\n                                    1       0.884615\n\n\n\n\n```python\ntt.tb\n```\n\n\n\n\n<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>AS PRESCRIBED OPIOID-1</th>\n      <th>HEROIN-1</th>\n      <th>HEROIN</th>\n      <th>pval</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.843494</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0.156506</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0.095652</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0.904348</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.882979</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0.117021</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0.115385</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0.884615</td>\n    </tr>\n  </tbody>\n</table>\n</div>\n\n\n\n```\nAS PRESCRIBED OPIOID-1\tHEROIN-1\tHEROIN\t\n0\t0\t0\t0.843494\n0\t0\t1\t0.156506\n1\t0\t0\t0.882979\n1\t0\t1\t0.117021\n```\n\n\n```python\n0.117021 / 0.156506\n```\n\n\n\n\n    0.7477093529960512\n\n\n\n\n```python\n\n```\n\n\n```python\nprob_of_heroin_given_presc_op = 0.359223\nprob_of_heroin_given_not_presc_op = 0.519213\n\nprob_of_heroin_given_presc_op / prob_of_heroin_given_not_presc_op\n```\n\n\n\n\n    0.6918605658949217\n\n\n\n\n```python\nprob_of_heroin_given_not_presc_op / prob_of_heroin_given_presc_op\n```\n\n\n\n\n    1.4453779407220584\n\n\n\n# Potential Calculus Experimentations\n\n\n```python\n# survey_dir = '/D/Dropbox/others/Miriam/python/ProcessedSurveys'\ndf_store = DfStore(survey_dir + '/{}.xlsx')\nlen(df_store)\n```\n\n\n\n\n    119\n\n\n\n\n```python\ncstore = VarSetCountsStore(df_store)\nv = mk_pvar_struct(df_store, only_for_cols_in_all_dfs=True)\ns = mk_pvar_str_struct(v)\nf, df = cstore.df_store.head()\ndf.head(3)\n```\n\n\n\n\n<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>category</th>\n      <th>RURAL</th>\n      <th>SUBURBAN</th>\n      <th>URBAN/CITY</th>\n      <th>HOMELESS</th>\n      <th>INCARCERATION</th>\n      <th>WORK</th>\n      <th>SON/DAUGHTER</th>\n      <th>SIBLING</th>\n      <th>FATHER/MOTHER</th>\n      <th>SPOUSE</th>\n      <th>...</th>\n      <th>HAL/LSD/XTC/CLUBDRUG</th>\n      <th>COCAINE/CRACK</th>\n      <th>METHAMPHETAMINE</th>\n      <th>AS PRESCRIBED OPIOID</th>\n      <th>NOT AS PRESCRIBED OPIOID</th>\n      <th>HEROIN</th>\n      <th>OTHER OPIOID</th>\n      <th>INJECTED</th>\n      <th>IN TREATMENT</th>\n      <th>Massachusetts</th>\n    </tr>\n    <tr>\n      <th>age</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      <th></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      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>16</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>17</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>18</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n  </tbody>\n</table>\n<p>3 rows \u00d7 29 columns</p>\n</div>\n\n\n\n\n```python\ncstore = VarSetCountsStore(df_store)\ncstore.mk_pvar_attrs()\n```\n\n\n```python\nfrom odus.dacc import DfStore, counts_of_kps, Dacc, plot_life_course, VarSetCountsStore, mk_pvar_struct, PotStore\npstore = PotStore(df_store)\npstore.mk_pvar_attrs()\np = pstore[v.homeless - 1, v.incarceration]\np\n```\n\n\n\n\n                              pval\n    HOMELESS-1 INCARCERATION      \n    0          0              1690\n               1               577\n    1          0               192\n               1                87\n\n\n\n\n```python\np / []\n```\n\n\n\n\n                                  pval\n    HOMELESS-1 INCARCERATION          \n    0          0              0.663786\n               1              0.226630\n    1          0              0.075412\n               1              0.034171\n\n\n\n\n```python\npstore[v.incarceration]\n```\n\n\n\n\n                   pval\n    INCARCERATION      \n    0              1989\n    1               676\n\n\n\n\n```python\npstore[v.alcohol-1, v.loss_of_loved_one]\n```\n\n\n\n\n                                 pval\n    ALCOHOL-1 LOSS OF LOVED ONE      \n    0         0                   990\n              1                    91\n    1         0                  1321\n              1                   144\n\n\n\n\n```python\ntw = pstore[v.tobacco, v.work]\nmw = pstore[v.marijuana, v.work]\naw = pstore[v.alcohol, v.work]\nw = pstore[v.work]\n\n```\n\n\n```python\nevid_t = Pot.from_hard_evidence(**{s.tobacco: 1})\nevid_m = Pot.from_hard_evidence(**{s.marijuana: 1})\nevid_a = Pot.from_hard_evidence(**{s.alcohol: 1})\nevid_a\n```\n\n\n\n\n             pval\n    ALCOHOL      \n    1           1\n\n\n\n\n```python\naw\n```\n\n\n\n\n                  pval\n    ALCOHOL WORK      \n    0       0      431\n            1      712\n    1       0      448\n            1     1074\n\n\n\n\n```python\nw / []\n```\n\n\n\n\n              pval\n    WORK          \n    0     0.329831\n    1     0.670169\n\n\n\n\n```python\n(evid_m * mw) / []\n```\n\n\n\n\n                        pval\n    MARIJUANA WORK          \n    1         0     0.350603\n              1     0.649397\n\n\n\n\n```python\n(evid_t * tw) / []\n```\n\n\n\n\n                      pval\n    TOBACCO WORK          \n    1       0     0.313001\n            1     0.686999\n\n\n\n\n```python\n(evid_a * aw) / []\n```\n\n\n\n\n                     pval\n    ALCOHOL WORK         \n    1       0     0.29435\n            1     0.70565\n\n\n\n# Extra scrap\n\n\n```python\n# from graphviz import Digraph\n# Digraph(body=\"\"\"\n# raw -> data -> count -> prob\n# raw [label=\"excel files (one per respondent)\" shape=folder]\n# data [label=\"dataframes\" shape=folder]\n# count [label=\"counts for any combinations of the variables in the data\" shape=box3d]\n# prob [label=\"probabilities for any combinations of the variables in the data\" shape=box3d]\n# \"\"\".split('\\n'))\n```\n\n\n```python\n\n```\n",
2020-08-03T14:45:25,595       "long_description_content_type": "text/markdown",
2020-08-03T14:45:25,595       "install_requires": [
2020-08-03T14:45:25,596         "py2store",
2020-08-03T14:45:25,596         "pandas",
2020-08-03T14:45:25,597         "numpy",
2020-08-03T14:45:25,598         "Pillow",
2020-08-03T14:45:25,598         "spyn",
2020-08-03T14:45:25,598         "matplotlib",
2020-08-03T14:45:25,599         "openpyxl",
2020-08-03T14:45:25,599         "argh"
2020-08-03T14:45:25,600       ],
2020-08-03T14:45:25,600       "description": "Tools to provide easy access to prepared data to data scientists that can't be asked.",
2020-08-03T14:45:25,601       "keywords": [
2020-08-03T14:45:25,601         "data",
2020-08-03T14:45:25,602         "data access",
2020-08-03T14:45:25,602         "drug use",
2020-08-03T14:45:25,603         "markov",
2020-08-03T14:45:25,603         "bayesian"
2020-08-03T14:45:25,603       ]
2020-08-03T14:45:25,937     }
2020-08-03T14:45:25,938     --------------------------------------------------------------------
2020-08-03T14:45:25,938     running egg_info
2020-08-03T14:45:25,943     creating /tmp/pip-pip-egg-info-k3z809f5/odus.egg-info
2020-08-03T14:45:25,960     writing /tmp/pip-pip-egg-info-k3z809f5/odus.egg-info/PKG-INFO
2020-08-03T14:45:25,970     writing dependency_links to /tmp/pip-pip-egg-info-k3z809f5/odus.egg-info/dependency_links.txt
2020-08-03T14:45:25,976     writing requirements to /tmp/pip-pip-egg-info-k3z809f5/odus.egg-info/requires.txt
2020-08-03T14:45:25,978     writing top-level names to /tmp/pip-pip-egg-info-k3z809f5/odus.egg-info/top_level.txt
2020-08-03T14:45:25,983     writing manifest file '/tmp/pip-pip-egg-info-k3z809f5/odus.egg-info/SOURCES.txt'
2020-08-03T14:45:25,990     reading manifest file '/tmp/pip-pip-egg-info-k3z809f5/odus.egg-info/SOURCES.txt'
2020-08-03T14:45:26,000     writing manifest file '/tmp/pip-pip-egg-info-k3z809f5/odus.egg-info/SOURCES.txt'
2020-08-03T14:45:26,122   Source in /tmp/pip-wheel-p7jn2a4e/odus has version 0.0.6, which satisfies requirement odus==0.0.6 from https://files.pythonhosted.org/packages/be/ea/a3d88b705ff73b7a7435335d1b2d6aa6202a78a9537d9ed219e4b77e428b/odus-0.0.6.tar.gz#sha256=e43b83b217218c592b938968f034b6f4045d5511baef54ec3c0b1bda60dbcad3
2020-08-03T14:45:26,124   Removed odus==0.0.6 from https://files.pythonhosted.org/packages/be/ea/a3d88b705ff73b7a7435335d1b2d6aa6202a78a9537d9ed219e4b77e428b/odus-0.0.6.tar.gz#sha256=e43b83b217218c592b938968f034b6f4045d5511baef54ec3c0b1bda60dbcad3 from build tracker '/tmp/pip-req-tracker-qi8cr40e'
2020-08-03T14:45:26,142 Building wheels for collected packages: odus
2020-08-03T14:45:26,152   Created temporary directory: /tmp/pip-wheel-1t5d36f_
2020-08-03T14:45:26,152   Building wheel for odus (setup.py): started
2020-08-03T14:45:26,153   Destination directory: /tmp/pip-wheel-1t5d36f_
2020-08-03T14:45:26,153   Running command /usr/bin/python3 -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-wheel-p7jn2a4e/odus/setup.py'"'"'; __file__='"'"'/tmp/pip-wheel-p7jn2a4e/odus/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' bdist_wheel -d /tmp/pip-wheel-1t5d36f_
2020-08-03T14:45:27,738   Setup params -------------------------------------------------------
2020-08-03T14:45:27,739   {
2020-08-03T14:45:27,740     "name": "odus",
2020-08-03T14:45:27,740     "version": "0.0.6",
2020-08-03T14:45:27,741     "url": "https://github.com/thorwhalen/odus",
2020-08-03T14:45:27,741     "author": "Thor Whalen",
2020-08-03T14:45:27,742     "author_email": "thorwhalen1@gmail.com",
2020-08-03T14:45:27,742     "license": "MIT",
2020-08-03T14:45:27,743     "include_package_data": true,
2020-08-03T14:45:27,743     "platforms": "any",
2020-08-03T14:45:27,744     "long_description": "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#Introduction\" data-toc-modified-id=\"Introduction-1\"><span class=\"toc-item-num\">1&nbsp;&nbsp;</span>Introduction</a></span></li><li><span><a href=\"#Installation\" data-toc-modified-id=\"Installation-2\"><span class=\"toc-item-num\">2&nbsp;&nbsp;</span>Installation</a></span></li><li><span><a href=\"#Getting-some-resources\" data-toc-modified-id=\"Getting-some-resources-3\"><span class=\"toc-item-num\">3&nbsp;&nbsp;</span>Getting some resources</a></span></li><li><span><a href=\"#Poking-around\" data-toc-modified-id=\"Poking-around-4\"><span class=\"toc-item-num\">4&nbsp;&nbsp;</span>Poking around</a></span><ul class=\"toc-item\"><li><span><a href=\"#df_store\" data-toc-modified-id=\"df_store-4.1\"><span class=\"toc-item-num\">4.1&nbsp;&nbsp;</span>df_store</a></span></li><li><span><a href=\"#Viewing-trajectories\" data-toc-modified-id=\"Viewing-trajectories-4.2\"><span class=\"toc-item-num\">4.2&nbsp;&nbsp;</span>Viewing trajectories</a></span><ul class=\"toc-item\"><li><span><a href=\"#Viewing-a-single-trajectory\" data-toc-modified-id=\"Viewing-a-single-trajectory-4.2.1\"><span class=\"toc-item-num\">4.2.1&nbsp;&nbsp;</span>Viewing a single trajectory</a></span></li><li><span><a href=\"#Flip-over-all-(or-some)-trajectories\" data-toc-modified-id=\"Flip-over-all-(or-some)-trajectories-4.2.2\"><span class=\"toc-item-num\">4.2.2&nbsp;&nbsp;</span>Flip over all (or some) trajectories</a></span></li></ul></li><li><span><a href=\"#Making-a-pdf-of-trajectories\" data-toc-modified-id=\"Making-a-pdf-of-trajectories-4.3\"><span class=\"toc-item-num\">4.3&nbsp;&nbsp;</span>Making a pdf of trajectories</a></span></li><li><span><a href=\"#Demo-s-and-v\" data-toc-modified-id=\"Demo-s-and-v-4.4\"><span class=\"toc-item-num\">4.4&nbsp;&nbsp;</span>Demo s and v</a></span></li><li><span><a href=\"#cstore\" data-toc-modified-id=\"cstore-4.5\"><span class=\"toc-item-num\">4.5&nbsp;&nbsp;</span>cstore</a></span></li><li><span><a href=\"#pstore\" data-toc-modified-id=\"pstore-4.6\"><span class=\"toc-item-num\">4.6&nbsp;&nbsp;</span>pstore</a></span></li><li><span><a href=\"#Scrap-place\" data-toc-modified-id=\"Scrap-place-4.7\"><span class=\"toc-item-num\">4.7&nbsp;&nbsp;</span>Scrap place</a></span></li></ul></li><li><span><a href=\"#Potential-Calculus-Experimentations\" data-toc-modified-id=\"Potential-Calculus-Experimentations-5\"><span class=\"toc-item-num\">5&nbsp;&nbsp;</span>Potential Calculus Experimentations</a></span></li><li><span><a href=\"#Extra-scrap\" data-toc-modified-id=\"Extra-scrap-6\"><span class=\"toc-item-num\">6&nbsp;&nbsp;</span>Extra scrap</a></span></li></ul></div>\n\n\n```python\n# %load_ext autoreload\n# %autoreload 2\n```\n\n# Introduction\n\nODUS (for Older Drug User Study) contains data and tools to study the drug use of older drug users.\n\nEssentially, there are these are tools:\n\n- To get prepared data on the 119 \"trajectories\" describing 31 variables (drug use, social, etc.) over time of 119 different respondents.\n\n- To vizualize these trajectories in various ways\n\n- To create pdfs of any selection of these trajectories and variables\n\n- To make count tables for any combinations of the variables: Essential step of any Markovian or Bayesian analysis.\n\n- To make probability (joint or conditional) tables from any combination of the variables\n\n- To operate on these count and probability tables, thus enabling inference operations\n\n\n# Installation\n\nYou need to have python 3.7+ to run this notebook.\n\nAnd you'll need to have `odus`, which you get by doing\n\n```\npip install odus\n```\n\n(And if you don't have pip then, well... how to put it... ha ha ha!)\n\nBut if you're the type, you can also just get the source from `https://github.com/thorwhalen/odus`. \n\nOh, and pull requests etc. are welcome!\n\nStars, likes, references, and coffee also welcome.\n\nAnd if you want to donate: Donate to a charity that will help the people understand and make policies surrounding the use of substances.\n\nA simple flowchart about the architecture:\n\n<img src=\"odus_store.png\" width=800 height=200>\n\n# Getting some resources\n\n\n```python\nfrom matplotlib.pylab import *\nfrom numpy import *\nimport seaborn as sns\n\nimport os\nfrom py2store.stores.local_store import RelativePathFormatStore\nfrom py2store.mixins import ReadOnlyMixin\nfrom py2store.base import Store\n\n\nfrom io import BytesIO\nfrom spyn.ppi.pot import Pot, ProbPot\nfrom collections import UserDict, Counter\nimport numpy as np\nimport pandas as pd\n\nfrom ut.ml.feature_extraction.sequential_var_sets import PVar, VarSet, DfData, VarSetFactory\nfrom IPython.display import Image\n\nfrom odus.analysis_utils import *\n\nfrom odus.dacc import DfStore, counts_of_kps, Dacc, VarSetCountsStore, \\\n    mk_pvar_struct, PotStore, _commun_columns_of_dfs, Struct, mk_pvar_str_struct, VarStr\n\nfrom odus.plot_utils import plot_life_course\n```\n\n\n```python\nfrom odus import data_dir, data_path_of\nsurvey_dir = data_dir\ndata_dir\n```\n\n\n\n\n    '/D/Dropbox/dev/p3/proj/odus/odus/data'\n\n\n\n\n```python\ndf_store = DfStore(data_dir + '/{}.xlsx')\nlen(df_store)\ncstore = VarSetCountsStore(df_store)\nv = mk_pvar_struct(df_store, only_for_cols_in_all_dfs=True)\ns = mk_pvar_str_struct(v)\nf, df = cstore.df_store.head()\npstore = PotStore(df_store)\n```\n\n# Poking around\n\n## df_store\n\nA df_store is a key-value store where the key is the xls file and the value is the prepared dataframe\n\n\n```python\nlen(df_store)\n```\n\n\n\n\n    119\n\n\n\n\n```python\nit = iter(df_store.values())\nfor i in range(5):  # skip five first\n    _ = next(it)\ndf = next(it)  # get the one I want\ndf.head(3)\n```\n\n\n\n\n<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>category</th>\n      <th>RURAL</th>\n      <th>SUBURBAN</th>\n      <th>URBAN/CITY</th>\n      <th>HOMELESS</th>\n      <th>INCARCERATION</th>\n      <th>WORK</th>\n      <th>SON/DAUGHTER</th>\n      <th>SIBLING</th>\n      <th>FATHER/MOTHER</th>\n      <th>SPOUSE</th>\n      <th>...</th>\n      <th>METHAMPHETAMINE</th>\n      <th>AS PRESCRIBED OPIOID</th>\n      <th>NOT AS PRESCRIBED OPIOID</th>\n      <th>HEROIN</th>\n      <th>OTHER OPIOID</th>\n      <th>INJECTED</th>\n      <th>IN TREATMENT</th>\n      <th>Selects States below</th>\n      <th>Georgia</th>\n      <th>Pennsylvania</th>\n    </tr>\n    <tr>\n      <th>age</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      <th></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      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>11</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>13</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n<p>3 rows \u00d7 31 columns</p>\n</div>\n\n\n\n\n```python\nprint(df.columns.values)\n```\n\n    ['RURAL' 'SUBURBAN' 'URBAN/CITY' 'HOMELESS' 'INCARCERATION' 'WORK'\n     'SON/DAUGHTER' 'SIBLING' 'FATHER/MOTHER' 'SPOUSE'\n     'OTHER (WHO?, FILL IN BRACKETS HERE)' 'FRIEND USER' 'FRIEND NON USER'\n     'MENTAL ILLNESS' 'PHYSICAL ILLNESS' 'LOSS OF LOVED ONE' 'TOBACCO'\n     'MARIJUANA' 'ALCOHOL' 'HAL/LSD/XTC/CLUBDRUG' 'COCAINE/CRACK'\n     'METHAMPHETAMINE' 'AS PRESCRIBED OPIOID' 'NOT AS PRESCRIBED OPIOID'\n     'HEROIN' 'OTHER OPIOID' 'INJECTED' 'IN TREATMENT' 'Selects States below'\n     'Georgia' 'Pennsylvania']\n\n\n\n```python\nt = df[['ALCOHOL', 'TOBACCO']]\nt.head(3)\n```\n\n\n\n\n<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>category</th>\n      <th>ALCOHOL</th>\n      <th>TOBACCO</th>\n    </tr>\n    <tr>\n      <th>age</th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>11</th>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>13</th>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n</div>\n\n\n\n\n```python\nc = Counter()\nfor i, r in t.iterrows():\n    c.update([tuple(r.to_list())])\nc\n```\n\n\n\n\n    Counter({(0, 0): 6, (1, 0): 4, (1, 1): 9, (0, 1): 2})\n\n\n\n\n```python\ndef count_tuples(dataframe):\n    c = Counter()\n    for i, r in dataframe.iterrows():\n        c.update([tuple(r.to_list())])\n    return c\n```\n\n\n```python\nfields = ['ALCOHOL', 'TOBACCO']\n# do it for every one\nc = Counter()\nfor df in df_store.values():\n    c.update(count_tuples(df[fields]))\nc\n```\n\n\n\n\n    Counter({(0, 1): 903, (1, 1): 1343, (0, 0): 240, (1, 0): 179})\n\n\n\n\n```python\npd.Series(c)\n```\n\n\n\n\n    0  1     903\n    1  1    1343\n    0  0     240\n    1  0     179\n    dtype: int64\n\n\n\n\n```python\n# Powerful! You can use that with several pairs and get some nice probabilities. Look up Naive Bayes.\n```\n\n## Viewing trajectories\n\n\n```python\nimport itertools\nfrom functools import partial\nfrom odus.util import write_images\nfrom odus.plot_utils import plot_life, life_plots, write_trajectories_to_file\n\nihead = lambda it: itertools.islice(it, 0, 5)\n```\n\n### Viewing a single trajectory\n\n\n```python\nk = next(iter(df_store))  # get the first key\nprint(f\"k: {k}\")  # print it\nplot_life(df_store[k])  # plot the trajectory\n```\n\n    k: surveys/B24.xlsx\n\n\n\n![png](img/output_26_1.png)\n\n\n\n```python\nplot_life(df_store[k], fields=[s.in_treatment, s.injected])  # only want two fields\n```\n\n\n![png](img/output_27_0.png)\n\n\n### Flip over all (or some) trajectories\n\n\n```python\ngen = life_plots(df_store)\n```\n\n\n```python\nnext(gen)  # launch to get the next trajectory\n```\n\n\n\n\n    <matplotlib.axes._subplots.AxesSubplot at 0x12b21f070>\n\n\n\n\n![png](img/output_30_1.png)\n\n\nGet three trajectories, but only over two fields.\n\n\n```python\n# fields = [s.in_treatment, s.injected]\nfields = [s.physical_illness, s.as_prescribed_opioid, s.heroin, s.other_opioid]\nkeys = list(df_store)[:10]\n# print(f\"keys={keys}\")\naxs = [x for x in life_plots(df_store, fields, keys=keys)];\n```\n\n\n![png](img/output_32_0.png)\n\n\n\n![png](img/output_32_1.png)\n\n\n\n![png](img/output_32_2.png)\n\n\n\n![png](img/output_32_3.png)\n\n\n\n![png](img/output_32_4.png)\n\n\n\n![png](img/output_32_5.png)\n\n\n\n![png](img/output_32_6.png)\n\n\n\n![png](img/output_32_7.png)\n\n\n\n![png](img/output_32_8.png)\n\n\n\n![png](img/output_32_9.png)\n\n\n## Making a pdf of trajectories\n\n\n```python\nwrite_trajectories_to_file(df_store, fields, keys, fp='three_respondents_two_fields.pdf');\n```\n\n\n```python\nwrite_trajectories_to_file(df_store, fp='all_respondents_all_fields.pdf');\n```\n\n\n```python\n \n```\n\n## Demo s and v\n\n\n```python\nprint(list(filter(lambda x: not x.startswith('__'), dir(s))))\n```\n\n    ['alcohol', 'as_prescribed_opioid', 'cocaine_crack', 'father_mother', 'hal_lsd_xtc_clubdrug', 'heroin', 'homeless', 'in_treatment', 'incarceration', 'injected', 'loss_of_loved_one', 'marijuana', 'mental_illness', 'methamphetamine', 'not_as_prescribed_opioid', 'other_opioid', 'physical_illness', 'rural', 'sibling', 'son_daughter', 'suburban', 'tobacco', 'urban_city', 'work']\n\n\n\n```python\ns.heroin\n```\n\n\n\n\n    'HEROIN'\n\n\n\n\n```python\nv.heroin\n```\n\n\n\n\n    PVar('HEROIN', 0)\n\n\n\n\n```python\nv.heroin - 1\n```\n\n\n\n\n    PVar('HEROIN', -1)\n\n\n\n## cstore\n\n\n```python\n# cstore[v.alcohol, v.tobacco]\ncstore[v.as_prescribed_opioid-1, v.heroin]\n```\n\n\n\n\n    Counter({(0, 0): 1026, (1, 0): 264, (0, 1): 1108, (1, 1): 148})\n\n\n\n\n```python\npd.Series(cstore[v.as_prescribed_opioid-1, v.heroin])\n```\n\n\n\n\n    0  0    1026\n    1  0     264\n    0  1    1108\n    1  1     148\n    dtype: int64\n\n\n\n\n```python\ncstore[v.alcohol, v.tobacco, v.heroin]\n```\n\n\n\n\n    Counter({(0, 0, 1): 427,\n             (1, 0, 1): 656,\n             (1, 1, 1): 687,\n             (0, 0, 0): 189,\n             (0, 1, 1): 476,\n             (0, 1, 0): 51,\n             (1, 0, 0): 133,\n             (1, 1, 0): 46})\n\n\n\n\n```python\ncstore[v.alcohol-1, v.alcohol]\n```\n\n\n\n\n    Counter({(0, 0): 994, (1, 1): 1375, (1, 0): 90, (0, 1): 87})\n\n\n\n\n```python\ncstore[v.alcohol-1, v.alcohol, v.tobacco]\n```\n\n\n\n\n    Counter({(0, 0, 1): 807,\n             (1, 1, 1): 1220,\n             (1, 0, 0): 26,\n             (0, 1, 1): 76,\n             (0, 0, 0): 187,\n             (1, 1, 0): 155,\n             (0, 1, 0): 11,\n             (1, 0, 1): 64})\n\n\n\n\n```python\nt = pd.Series(cstore[v.alcohol-1, v.alcohol, v.tobacco])\nt.loc[t.index]\n```\n\n\n\n\n    <pandas.core.indexing._LocIndexer at 0x130955db0>\n\n\n\n## pstore\n\n\n```python\nt = pstore[s.alcohol-1, s.alcohol]\nt\n```\n\n\n\n\n                       pval\n    ALCOHOL-1 ALCOHOL      \n    0         0         994\n              1          87\n    1         0          90\n              1        1375\n\n\n\n\n```python\nt.tb\n```\n\n\n\n\n<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>ALCOHOL-1</th>\n      <th>ALCOHOL</th>\n      <th>pval</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>0</td>\n      <td>994</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>1</td>\n      <td>87</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>0</td>\n      <td>90</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>1</td>\n      <td>1375</td>\n    </tr>\n  </tbody>\n</table>\n</div>\n\n\n\n\n```python\nt / []\n```\n\n\n\n\n                           pval\n    ALCOHOL-1 ALCOHOL          \n    0         0        0.390416\n              1        0.034171\n    1         0        0.035350\n              1        0.540063\n\n\n\n\n```python\nt[s.alcohol-1]\n```\n\n\n\n\n               pval\n    ALCOHOL-1      \n    0          1081\n    1          1465\n\n\n\n\n```python\nt / t[s.alcohol-1]  # cond prob!\n```\n\n\n\n\n                           pval\n    ALCOHOL-1 ALCOHOL          \n    0         0        0.919519\n              1        0.080481\n    1         0        0.061433\n              1        0.938567\n\n\n\n\n```python\ntt = pstore[s.alcohol, s.tobacco]\ntt\n```\n\n\n\n\n                     pval\n    ALCOHOL TOBACCO      \n    0       0         240\n            1         903\n    1       0         179\n            1        1343\n\n\n\n\n```python\ntt / tt[s.alcohol]\n```\n\n\n\n\n                         pval\n    ALCOHOL TOBACCO          \n    0       0        0.209974\n            1        0.790026\n    1       0        0.117608\n            1        0.882392\n\n\n\n\n```python\ntt / tt[s.tobacco]\n```\n\n\n\n\n                         pval\n    ALCOHOL TOBACCO          \n    0       0        0.572792\n    1       0        0.427208\n    0       1        0.402048\n    1       1        0.597952\n\n\n\n\n```python\n\n```\n\n## Scrap place\n\n\n```python\nt = pstore[s.as_prescribed_opioid-1, s.heroin-1, s.heroin]\nt\n\n```\n\n\n\n\n                                            pval\n    AS PRESCRIBED OPIOID-1 HEROIN-1 HEROIN      \n    0                      0        0        927\n                                    1        172\n                           1        0         99\n                                    1        936\n    1                      0        0        249\n                                    1         33\n                           1        0         15\n                                    1        115\n\n\n\n\n```python\ntt = t / t[s.as_prescribed_opioid-1, s.heroin-1]  # cond prob!\ntt\n```\n\n\n\n\n                                                pval\n    AS PRESCRIBED OPIOID-1 HEROIN-1 HEROIN          \n    0                      0        0       0.843494\n                                    1       0.156506\n                           1        0       0.095652\n                                    1       0.904348\n    1                      0        0       0.882979\n                                    1       0.117021\n                           1        0       0.115385\n                                    1       0.884615\n\n\n\n\n```python\ntt.tb\n```\n\n\n\n\n<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>AS PRESCRIBED OPIOID-1</th>\n      <th>HEROIN-1</th>\n      <th>HEROIN</th>\n      <th>pval</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.843494</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0.156506</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0.095652</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0.904348</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.882979</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0.117021</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0.115385</td>\n    </tr>\n    <tr>\n      <th></th>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0.884615</td>\n    </tr>\n  </tbody>\n</table>\n</div>\n\n\n\n```\nAS PRESCRIBED OPIOID-1\tHEROIN-1\tHEROIN\t\n0\t0\t0\t0.843494\n0\t0\t1\t0.156506\n1\t0\t0\t0.882979\n1\t0\t1\t0.117021\n```\n\n\n```python\n0.117021 / 0.156506\n```\n\n\n\n\n    0.7477093529960512\n\n\n\n\n```python\n\n```\n\n\n```python\nprob_of_heroin_given_presc_op = 0.359223\nprob_of_heroin_given_not_presc_op = 0.519213\n\nprob_of_heroin_given_presc_op / prob_of_heroin_given_not_presc_op\n```\n\n\n\n\n    0.6918605658949217\n\n\n\n\n```python\nprob_of_heroin_given_not_presc_op / prob_of_heroin_given_presc_op\n```\n\n\n\n\n    1.4453779407220584\n\n\n\n# Potential Calculus Experimentations\n\n\n```python\n# survey_dir = '/D/Dropbox/others/Miriam/python/ProcessedSurveys'\ndf_store = DfStore(survey_dir + '/{}.xlsx')\nlen(df_store)\n```\n\n\n\n\n    119\n\n\n\n\n```python\ncstore = VarSetCountsStore(df_store)\nv = mk_pvar_struct(df_store, only_for_cols_in_all_dfs=True)\ns = mk_pvar_str_struct(v)\nf, df = cstore.df_store.head()\ndf.head(3)\n```\n\n\n\n\n<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>category</th>\n      <th>RURAL</th>\n      <th>SUBURBAN</th>\n      <th>URBAN/CITY</th>\n      <th>HOMELESS</th>\n      <th>INCARCERATION</th>\n      <th>WORK</th>\n      <th>SON/DAUGHTER</th>\n      <th>SIBLING</th>\n      <th>FATHER/MOTHER</th>\n      <th>SPOUSE</th>\n      <th>...</th>\n      <th>HAL/LSD/XTC/CLUBDRUG</th>\n      <th>COCAINE/CRACK</th>\n      <th>METHAMPHETAMINE</th>\n      <th>AS PRESCRIBED OPIOID</th>\n      <th>NOT AS PRESCRIBED OPIOID</th>\n      <th>HEROIN</th>\n      <th>OTHER OPIOID</th>\n      <th>INJECTED</th>\n      <th>IN TREATMENT</th>\n      <th>Massachusetts</th>\n    </tr>\n    <tr>\n      <th>age</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      <th></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      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>16</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>17</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>18</th>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n  </tbody>\n</table>\n<p>3 rows \u00d7 29 columns</p>\n</div>\n\n\n\n\n```python\ncstore = VarSetCountsStore(df_store)\ncstore.mk_pvar_attrs()\n```\n\n\n```python\nfrom odus.dacc import DfStore, counts_of_kps, Dacc, plot_life_course, VarSetCountsStore, mk_pvar_struct, PotStore\npstore = PotStore(df_store)\npstore.mk_pvar_attrs()\np = pstore[v.homeless - 1, v.incarceration]\np\n```\n\n\n\n\n                              pval\n    HOMELESS-1 INCARCERATION      \n    0          0              1690\n               1               577\n    1          0               192\n               1                87\n\n\n\n\n```python\np / []\n```\n\n\n\n\n                                  pval\n    HOMELESS-1 INCARCERATION          \n    0          0              0.663786\n               1              0.226630\n    1          0              0.075412\n               1              0.034171\n\n\n\n\n```python\npstore[v.incarceration]\n```\n\n\n\n\n                   pval\n    INCARCERATION      \n    0              1989\n    1               676\n\n\n\n\n```python\npstore[v.alcohol-1, v.loss_of_loved_one]\n```\n\n\n\n\n                                 pval\n    ALCOHOL-1 LOSS OF LOVED ONE      \n    0         0                   990\n              1                    91\n    1         0                  1321\n              1                   144\n\n\n\n\n```python\ntw = pstore[v.tobacco, v.work]\nmw = pstore[v.marijuana, v.work]\naw = pstore[v.alcohol, v.work]\nw = pstore[v.work]\n\n```\n\n\n```python\nevid_t = Pot.from_hard_evidence(**{s.tobacco: 1})\nevid_m = Pot.from_hard_evidence(**{s.marijuana: 1})\nevid_a = Pot.from_hard_evidence(**{s.alcohol: 1})\nevid_a\n```\n\n\n\n\n             pval\n    ALCOHOL      \n    1           1\n\n\n\n\n```python\naw\n```\n\n\n\n\n                  pval\n    ALCOHOL WORK      \n    0       0      431\n            1      712\n    1       0      448\n            1     1074\n\n\n\n\n```python\nw / []\n```\n\n\n\n\n              pval\n    WORK          \n    0     0.329831\n    1     0.670169\n\n\n\n\n```python\n(evid_m * mw) / []\n```\n\n\n\n\n                        pval\n    MARIJUANA WORK          \n    1         0     0.350603\n              1     0.649397\n\n\n\n\n```python\n(evid_t * tw) / []\n```\n\n\n\n\n                      pval\n    TOBACCO WORK          \n    1       0     0.313001\n            1     0.686999\n\n\n\n\n```python\n(evid_a * aw) / []\n```\n\n\n\n\n                     pval\n    ALCOHOL WORK         \n    1       0     0.29435\n            1     0.70565\n\n\n\n# Extra scrap\n\n\n```python\n# from graphviz import Digraph\n# Digraph(body=\"\"\"\n# raw -> data -> count -> prob\n# raw [label=\"excel files (one per respondent)\" shape=folder]\n# data [label=\"dataframes\" shape=folder]\n# count [label=\"counts for any combinations of the variables in the data\" shape=box3d]\n# prob [label=\"probabilities for any combinations of the variables in the data\" shape=box3d]\n# \"\"\".split('\\n'))\n```\n\n\n```python\n\n```\n",
2020-08-03T14:45:27,746     "long_description_content_type": "text/markdown",
2020-08-03T14:45:27,746     "install_requires": [
2020-08-03T14:45:27,747       "py2store",
2020-08-03T14:45:27,748       "pandas",
2020-08-03T14:45:27,748       "numpy",
2020-08-03T14:45:27,749       "Pillow",
2020-08-03T14:45:27,749       "spyn",
2020-08-03T14:45:27,750       "matplotlib",
2020-08-03T14:45:27,751       "openpyxl",
2020-08-03T14:45:27,751       "argh"
2020-08-03T14:45:27,752     ],
2020-08-03T14:45:27,752     "description": "Tools to provide easy access to prepared data to data scientists that can't be asked.",
2020-08-03T14:45:27,753     "keywords": [
2020-08-03T14:45:27,753       "data",
2020-08-03T14:45:27,754       "data access",
2020-08-03T14:45:27,754       "drug use",
2020-08-03T14:45:27,755       "markov",
2020-08-03T14:45:27,756       "bayesian"
2020-08-03T14:45:27,757     ]
2020-08-03T14:45:27,757   }
2020-08-03T14:45:27,758   --------------------------------------------------------------------
2020-08-03T14:45:28,081   running bdist_wheel
2020-08-03T14:45:28,094   running build
2020-08-03T14:45:28,105   installing to build/bdist.linux-armv7l/wheel
2020-08-03T14:45:28,106   running install
2020-08-03T14:45:28,108   running install_egg_info
2020-08-03T14:45:28,150   running egg_info
2020-08-03T14:45:28,166   writing odus.egg-info/PKG-INFO
2020-08-03T14:45:28,174   writing dependency_links to odus.egg-info/dependency_links.txt
2020-08-03T14:45:28,178   writing requirements to odus.egg-info/requires.txt
2020-08-03T14:45:28,180   writing top-level names to odus.egg-info/top_level.txt
2020-08-03T14:45:28,188   reading manifest file 'odus.egg-info/SOURCES.txt'
2020-08-03T14:45:28,195   writing manifest file 'odus.egg-info/SOURCES.txt'
2020-08-03T14:45:28,200   Copying odus.egg-info to build/bdist.linux-armv7l/wheel/odus-0.0.6-py3.7.egg-info
2020-08-03T14:45:28,228   running install_scripts
2020-08-03T14:45:28,440   creating build/bdist.linux-armv7l/wheel/odus-0.0.6.dist-info/WHEEL
2020-08-03T14:45:28,445   creating '/tmp/pip-wheel-1t5d36f_/odus-0.0.6-py3-none-any.whl' and adding 'build/bdist.linux-armv7l/wheel' to it
2020-08-03T14:45:28,456   adding 'odus-0.0.6.dist-info/METADATA'
2020-08-03T14:45:28,460   adding 'odus-0.0.6.dist-info/WHEEL'
2020-08-03T14:45:28,462   adding 'odus-0.0.6.dist-info/top_level.txt'
2020-08-03T14:45:28,464   adding 'odus-0.0.6.dist-info/RECORD'
2020-08-03T14:45:28,466   removing build/bdist.linux-armv7l/wheel
2020-08-03T14:45:28,592   Building wheel for odus (setup.py): finished with status 'done'
2020-08-03T14:45:28,596   Created wheel for odus: filename=odus-0.0.6-py3-none-any.whl size=6690 sha256=1e9ed08a2db3bd732081b27fc0f51b1663c70f2f9d791af3cca9d7393d8bb542
2020-08-03T14:45:28,597   Stored in directory: /tmp/pip-ephem-wheel-cache-c2ro9x8m/wheels/76/19/75/cb0c81022e84cb532d1f95dbb271f65932072bfdc33f672867
2020-08-03T14:45:28,601 Successfully built odus
2020-08-03T14:45:28,608 Removed build tracker: '/tmp/pip-req-tracker-qi8cr40e'