{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "%matplotlib inline\n", "\n", "import itertools\n", "import pathlib\n", "import warnings\n", "\n", "import matplotlib.pyplot as plt\n", "import pandas as pd\n", "import scipy.stats\n", "\n", "def fisher_exact(x, /, *, func=scipy.stats.fisher_exact) -> pd.Series:\n", " return pd.Series(func(x), index=('odds', 'p'))\n", "\n", "def pearsonr(df, /, left, right, *, func=scipy.stats.pearsonr) -> pd.Series:\n", " df = df[[left, right]].dropna()\n", " name = f'{left} & {right}'\n", " with warnings.catch_warnings():\n", " warnings.simplefilter('ignore', scipy.stats.ConstantInputWarning)\n", " result = func(df[left], df[right])\n", " return pd.Series(result, index=('r', 'p'), name=name)\n", "\n", "plt.style.use('classic')\n", "plt.rcParams.update({'figure.figsize': (6, 4), 'figure.facecolor': 'w',\n", " 'figure.subplot.bottom': .125, 'font.size': 10, 'savefig.dpi': 72})" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "RangeIndex: 14773 entries, 0 to 14772\n", "Data columns (total 4 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 Language 14773 non-null string\n", " 1 Discharger 14773 non-null string\n", " 2 Cell 14773 non-null string\n", " 3 Form 7969 non-null string\n", "dtypes: string(4)\n", "memory usage: 461.8 KB\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
LanguageDischargerCellForm
0AinuNone1sku
1AinuNone1pas
2AinuNone2se
3AinuNone2peci
4AinuNone3s<NA>
5AinuNone3p<NA>
6AinuNonexan
7AinuNone1s->2seci
8AinuNone1s->2peci
9AinuNone1s->3sku
\n", "
" ], "text/plain": [ " Language Discharger Cell Form\n", "0 Ainu None 1s ku\n", "1 Ainu None 1p as\n", "2 Ainu None 2s e\n", "3 Ainu None 2p eci\n", "4 Ainu None 3s \n", "5 Ainu None 3p \n", "6 Ainu None x an\n", "7 Ainu None 1s->2s eci\n", "8 Ainu None 1s->2p eci\n", "9 Ainu None 1s->3s ku" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "CSV = pathlib.Path('nonportmanteau.csv')\n", "\n", "RENAME = {'Quechua (Ayacucho)': 'Ayacucho',\n", " 'Tlachichilco Tepehuan': 'Tepehua',\n", " 'Lakhota': 'Lakota'}\n", "\n", "tb = (pd.read_csv(CSV, encoding='utf-8', na_values='', keep_default_na=False,\n", " dtype={'Language': 'string',\n", " 'Discharger': 'string',\n", " 'Cell': 'string',\n", " 'Form': 'string'})\n", " .assign(Language=lambda x: x['Language'].replace(RENAME),\n", " Discharger=lambda x: (x['Discharger']\n", " .str.replace('Fifty', '50')\n", " .str.replace('Sixty', '60')\n", " .str.replace('Hundred', '/100')\n", " .str.replace('Discharger', ''))))\n", "\n", "tb.info()\n", "assert tb.set_index(['Language', 'Discharger', 'Cell']).index.is_unique\n", "tb.head(10)" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
LanguageCellForm
Discharger
All50/10013431343350
All6013431343143
All60/10013431343496
AllIntransPerfect13431343615
AllPerfect134313431050
Free50/10013431343589
Free6013431343434
Free60/10013431343828
FreePerfect134313431133
Intrans134313431014
None134313431317
\n", "
" ], "text/plain": [ " Language Cell Form\n", "Discharger \n", "All50/100 1343 1343 350\n", "All60 1343 1343 143\n", "All60/100 1343 1343 496\n", "AllIntransPerfect 1343 1343 615\n", "AllPerfect 1343 1343 1050\n", "Free50/100 1343 1343 589\n", "Free60 1343 1343 434\n", "Free60/100 1343 1343 828\n", "FreePerfect 1343 1343 1133\n", "Intrans 1343 1343 1014\n", "None 1343 1343 1317" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "tb.groupby('Discharger').count()" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
LanguageDischargerCellTransLocalDirectForm
0AinuNone1sFalseFalseTrueku
1AinuNone1pFalseFalseTrueas
2AinuNone2sFalseFalseTruee
3AinuNone2pFalseFalseTrueeci
4AinuNone3sFalseFalseTrue<NA>
5AinuNone3pFalseFalseTrue<NA>
6AinuNonexFalseFalseTruean
7AinuNone1s->2sTrueTrueTrueeci
8AinuNone1s->2pTrueTrueTrueeci
9AinuNone1s->3sTrueFalseTrueku
\n", "
" ], "text/plain": [ " Language Discharger Cell Trans Local Direct Form\n", "0 Ainu None 1s False False True ku\n", "1 Ainu None 1p False False True as\n", "2 Ainu None 2s False False True e\n", "3 Ainu None 2p False False True eci\n", "4 Ainu None 3s False False True \n", "5 Ainu None 3p False False True \n", "6 Ainu None x False False True an\n", "7 Ainu None 1s->2s True True True eci\n", "8 Ainu None 1s->2p True True True eci\n", "9 Ainu None 1s->3s True False True ku" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "SEP = '->'\n", "\n", "def is_local(cellkey, /, *, sep=SEP, persons=('1', '2')) -> bool:\n", " (subj, _, obj) = cellkey.partition(sep)\n", " return any(p in subj for p in persons) and any(p in obj for p in persons)\n", "\n", "def is_inverse(cellkey, /, *, sep=SEP, hierarchy=('1', '2', '3', '4', 'x')) -> bool:\n", " (subj, _, obj) = cellkey.partition(sep)\n", " for high, low in itertools.combinations(hierarchy, 2):\n", " if low in subj and high in obj:\n", " return True\n", " return False\n", "\n", "tb = (tb.assign(Trans=lambda x: x['Cell'].str.contains(SEP).astype('bool'),\n", " Local=lambda x: x['Cell'].map(is_local),\n", " Direct=lambda x: ~x['Cell'].map(is_inverse))\n", " [['Language',\n", " 'Discharger',\n", " 'Cell', 'Trans', 'Local', 'Direct',\n", " 'Form']])\n", "\n", "tb.head(10)" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Discharger
0None
1Intrans
2FreePerfect
3Free60/100
4Free50/100
5Free60
6AllPerfect
7AllIntransPerfect
8All60/100
9All50/100
10All60
\n", "
" ], "text/plain": [ " Discharger\n", "0 None\n", "1 Intrans\n", "2 FreePerfect\n", "3 Free60/100\n", "4 Free50/100\n", "5 Free60\n", "6 AllPerfect\n", "7 AllIntransPerfect\n", "8 All60/100\n", "9 All50/100\n", "10 All60" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "disc = tb['Discharger'].unique()\n", "\n", "pd.DataFrame(disc, columns=['Discharger'])" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DischargerNoneIntransFreePerfectFree60/100Free50/100...AllPerfectAllIntransPerfectAll60/100All50/100All60
LocalFalseTrueFalseTrueFalseTrueFalseTrueFalseTrue...FalseTrueFalseTrueFalseTrueFalseTrueFalseTrue
Language
Ainu29819425619690...2561501969000
Aleut451827039182615206...39182702615206143
Ayacucho2481081989696...19746444404
Bella Coola2081561883535...188145353510
Chuckchi1981941985656...198194565634
Darai208142162162162...142142000000
Fox328328328308248...3283281480800
Hixkaryana22682206164164...2068216416480
Jaqaru7333634343...4320111111
Jumjum288188248228228...24816812512552
Karuk20896197155115...197851447220
Ket848748768528528...640100200100100
Kunama5718471652184212284...328124306284270
Lakota2381361870202...18744020200
Maricopa168481684400...16848440000
Maung968808968917631...958808220100200
Mordvin208207208112112...14518710210281
Nocte208128184122122...1848412212262
Reyesano208001600000...16000000000
Sahu968796545525365...54533052536500
Siuslawan571857185318531862...47115718000000
Tepehua238178228148108...22817814810806
Thangmi2081021866161...18680614101
Turkana2082220814862...20822404040
Wardaman278208238208157...218148644210
Yimas45184518451845184518...43183911341630936
\n", "

26 rows × 22 columns

\n", "
" ], "text/plain": [ "Discharger None Intrans FreePerfect Free60/100 \\\n", "Local False True False True False True False True \n", "Language \n", "Ainu 29 8 19 4 25 6 19 6 \n", "Aleut 45 18 27 0 39 18 26 15 \n", "Ayacucho 24 8 10 8 19 8 9 6 \n", "Bella Coola 20 8 15 6 18 8 3 5 \n", "Chuckchi 19 8 19 4 19 8 5 6 \n", "Darai 20 8 14 2 16 2 16 2 \n", "Fox 32 8 32 8 32 8 30 8 \n", "Hixkaryana 22 6 8 2 20 6 16 4 \n", "Jaqaru 7 3 3 3 6 3 4 3 \n", "Jumjum 28 8 18 8 24 8 22 8 \n", "Karuk 20 8 9 6 19 7 15 5 \n", "Ket 84 8 74 8 76 8 52 8 \n", "Kunama 57 18 47 16 52 18 42 12 \n", "Lakota 23 8 13 6 18 7 0 2 \n", "Maricopa 16 8 4 8 16 8 4 4 \n", "Maung 96 8 80 8 96 8 91 7 \n", "Mordvin 20 8 20 7 20 8 11 2 \n", "Nocte 20 8 12 8 18 4 12 2 \n", "Reyesano 20 8 0 0 16 0 0 0 \n", "Sahu 96 8 79 6 54 5 52 5 \n", "Siuslawan 57 18 57 18 53 18 53 18 \n", "Tepehua 23 8 17 8 22 8 14 8 \n", "Thangmi 20 8 10 2 18 6 6 1 \n", "Turkana 20 8 2 2 20 8 14 8 \n", "Wardaman 27 8 20 8 23 8 20 8 \n", "Yimas 45 18 45 18 45 18 45 18 \n", "\n", "Discharger Free50/100 ... AllPerfect AllIntransPerfect \\\n", "Local False True ... False True False True \n", "Language ... \n", "Ainu 9 0 ... 25 6 15 0 \n", "Aleut 20 6 ... 39 18 27 0 \n", "Ayacucho 9 6 ... 19 7 4 6 \n", "Bella Coola 3 5 ... 18 8 14 5 \n", "Chuckchi 5 6 ... 19 8 19 4 \n", "Darai 16 2 ... 14 2 14 2 \n", "Fox 24 8 ... 32 8 32 8 \n", "Hixkaryana 16 4 ... 20 6 8 2 \n", "Jaqaru 4 3 ... 4 3 2 0 \n", "Jumjum 22 8 ... 24 8 16 8 \n", "Karuk 11 5 ... 19 7 8 5 \n", "Ket 52 8 ... 64 0 10 0 \n", "Kunama 28 4 ... 32 8 12 4 \n", "Lakota 0 2 ... 18 7 4 4 \n", "Maricopa 0 0 ... 16 8 4 8 \n", "Maung 63 1 ... 95 8 80 8 \n", "Mordvin 11 2 ... 14 5 18 7 \n", "Nocte 12 2 ... 18 4 8 4 \n", "Reyesano 0 0 ... 16 0 0 0 \n", "Sahu 36 5 ... 54 5 33 0 \n", "Siuslawan 6 2 ... 47 11 57 18 \n", "Tepehua 10 8 ... 22 8 17 8 \n", "Thangmi 6 1 ... 18 6 8 0 \n", "Turkana 6 2 ... 20 8 2 2 \n", "Wardaman 15 7 ... 21 8 14 8 \n", "Yimas 45 18 ... 43 18 39 11 \n", "\n", "Discharger All60/100 All50/100 All60 \n", "Local False True False True False True \n", "Language \n", "Ainu 19 6 9 0 0 0 \n", "Aleut 26 15 20 6 14 3 \n", "Ayacucho 4 4 4 4 0 4 \n", "Bella Coola 3 5 3 5 1 0 \n", "Chuckchi 5 6 5 6 3 4 \n", "Darai 0 0 0 0 0 0 \n", "Fox 14 8 0 8 0 0 \n", "Hixkaryana 16 4 16 4 8 0 \n", "Jaqaru 1 1 1 1 1 1 \n", "Jumjum 12 5 12 5 5 2 \n", "Karuk 14 4 7 2 2 0 \n", "Ket 20 0 10 0 10 0 \n", "Kunama 30 6 28 4 27 0 \n", "Lakota 0 2 0 2 0 0 \n", "Maricopa 4 4 0 0 0 0 \n", "Maung 22 0 10 0 20 0 \n", "Mordvin 10 2 10 2 8 1 \n", "Nocte 12 2 12 2 6 2 \n", "Reyesano 0 0 0 0 0 0 \n", "Sahu 52 5 36 5 0 0 \n", "Siuslawan 0 0 0 0 0 0 \n", "Tepehua 14 8 10 8 0 6 \n", "Thangmi 6 1 4 1 0 1 \n", "Turkana 4 0 4 0 4 0 \n", "Wardaman 6 4 4 2 1 0 \n", "Yimas 34 16 30 9 3 6 \n", "\n", "[26 rows x 22 columns]" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ff = (tb[tb['Trans']]\n", " .pivot_table('Form', 'Language', ['Discharger', 'Local'],\n", " aggfunc='count', fill_value=0))\n", "\n", "ff[disc]" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
LocalFalseTrue
Discharger
None890241
Intrans654174
FreePerfect784214
Free60/100581171
Free50/100429115
Free6034094
AllPerfect731185
AllIntransPerfect465122
All60/100328108
All50/10023576
All6011330
\n", "
" ], "text/plain": [ "Local False True \n", "Discharger \n", "None 890 241\n", "Intrans 654 174\n", "FreePerfect 784 214\n", "Free60/100 581 171\n", "Free50/100 429 115\n", "Free60 340 94\n", "AllPerfect 731 185\n", "AllIntransPerfect 465 122\n", "All60/100 328 108\n", "All50/100 235 76\n", "All60 113 30" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ff.sum().unstack('Local').loc[disc]" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
LocalFalseTrue
Discharger
None100.000000100.000000
Intrans64.86060172.489316
FreePerfect89.59280187.019231
Free60/10058.14522268.429487
Free50/10045.05573050.320513
Free6034.66263139.583333
AllPerfect84.15569877.617521
AllIntransPerfect50.61374350.160256
All60/10035.46420743.482906
All50/10026.23982732.905983
All6011.95707612.820513
\n", "
" ], "text/plain": [ "Local False True \n", "Discharger \n", "None 100.000000 100.000000\n", "Intrans 64.860601 72.489316\n", "FreePerfect 89.592801 87.019231\n", "Free60/100 58.145222 68.429487\n", "Free50/100 45.055730 50.320513\n", "Free60 34.662631 39.583333\n", "AllPerfect 84.155698 77.617521\n", "AllIntransPerfect 50.613743 50.160256\n", "All60/100 35.464207 43.482906\n", "All50/100 26.239827 32.905983\n", "All60 11.957076 12.820513" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df = 100 * ff / ff['None']\n", "\n", "nf = df / df.T.groupby(level='Discharger').sum().T\n", "\n", "xf = df.join(nf, rsuffix=' (norm)')\n", "\n", "df.mean().unstack('Local').loc[disc]" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DischargerNoneIntransFreePerfectFree60/100Free50/100...AllPerfectAllIntransPerfectAll60/100All50/100All60
LocalFalseTrueFalseTrueFalseTrueFalseTrueFalseTrue...FalseTrueFalseTrueFalseTrueFalseTrueFalseTrue
Language
Ainu100100665086756675310...8675520667531000
Aleut1001006008710058834433...87100600588344333117
Ayacucho100100421007910038753875...7988177517501750050
Bella Coola10010075759010015621562...9010070621562156250
Chuckchi1001001005010010026752675...10010010050267526751650
Darai1001007025802580258025...70257025000000
Fox1001001001001001009410075100...10010010010044100010000
Hixkaryana10010036339110073677367...91100363373677367360
Jaqaru10010043100861005710057100...57100290143314331433
Jumjum10010064100861007910079100...8610057100436243621825
Karuk1001004575958875625562...9588406270503525100
Ket10010088100901006210062100...760120240120120
Kunama10010082899110074674922...5644212253334922470
Lakota10010057757888025025...7888175002502500
Maricopa10010025100100100255000...1001002510025500000
Maung1001008310010010095886612...9910083100230100210
Mordvin1001001008810010055255525...70629088502550254012
Nocte10010060100905060256025...90504050602560253025
Reyesano100100008000000...80000000000
Sahu1001008275566254623862...56623405462386200
Siuslawan10010010010093100931001111...8261100100000000
Tepehua10010074100961006110043100...96100741006110043100075
Thangmi1001005025907530123012...907540030122012012
Turkana1001001025100100701003025...1001001025200200200
Wardaman1001007410085100741005688...78100521002250152540
Yimas100100100100100100100100100100...96100876176896750733
\n", "

26 rows × 22 columns

\n", "
" ], "text/plain": [ "Discharger None Intrans FreePerfect Free60/100 \\\n", "Local False True False True False True False True \n", "Language \n", "Ainu 100 100 66 50 86 75 66 75 \n", "Aleut 100 100 60 0 87 100 58 83 \n", "Ayacucho 100 100 42 100 79 100 38 75 \n", "Bella Coola 100 100 75 75 90 100 15 62 \n", "Chuckchi 100 100 100 50 100 100 26 75 \n", "Darai 100 100 70 25 80 25 80 25 \n", "Fox 100 100 100 100 100 100 94 100 \n", "Hixkaryana 100 100 36 33 91 100 73 67 \n", "Jaqaru 100 100 43 100 86 100 57 100 \n", "Jumjum 100 100 64 100 86 100 79 100 \n", "Karuk 100 100 45 75 95 88 75 62 \n", "Ket 100 100 88 100 90 100 62 100 \n", "Kunama 100 100 82 89 91 100 74 67 \n", "Lakota 100 100 57 75 78 88 0 25 \n", "Maricopa 100 100 25 100 100 100 25 50 \n", "Maung 100 100 83 100 100 100 95 88 \n", "Mordvin 100 100 100 88 100 100 55 25 \n", "Nocte 100 100 60 100 90 50 60 25 \n", "Reyesano 100 100 0 0 80 0 0 0 \n", "Sahu 100 100 82 75 56 62 54 62 \n", "Siuslawan 100 100 100 100 93 100 93 100 \n", "Tepehua 100 100 74 100 96 100 61 100 \n", "Thangmi 100 100 50 25 90 75 30 12 \n", "Turkana 100 100 10 25 100 100 70 100 \n", "Wardaman 100 100 74 100 85 100 74 100 \n", "Yimas 100 100 100 100 100 100 100 100 \n", "\n", "Discharger Free50/100 ... AllPerfect AllIntransPerfect \\\n", "Local False True ... False True False True \n", "Language ... \n", "Ainu 31 0 ... 86 75 52 0 \n", "Aleut 44 33 ... 87 100 60 0 \n", "Ayacucho 38 75 ... 79 88 17 75 \n", "Bella Coola 15 62 ... 90 100 70 62 \n", "Chuckchi 26 75 ... 100 100 100 50 \n", "Darai 80 25 ... 70 25 70 25 \n", "Fox 75 100 ... 100 100 100 100 \n", "Hixkaryana 73 67 ... 91 100 36 33 \n", "Jaqaru 57 100 ... 57 100 29 0 \n", "Jumjum 79 100 ... 86 100 57 100 \n", "Karuk 55 62 ... 95 88 40 62 \n", "Ket 62 100 ... 76 0 12 0 \n", "Kunama 49 22 ... 56 44 21 22 \n", "Lakota 0 25 ... 78 88 17 50 \n", "Maricopa 0 0 ... 100 100 25 100 \n", "Maung 66 12 ... 99 100 83 100 \n", "Mordvin 55 25 ... 70 62 90 88 \n", "Nocte 60 25 ... 90 50 40 50 \n", "Reyesano 0 0 ... 80 0 0 0 \n", "Sahu 38 62 ... 56 62 34 0 \n", "Siuslawan 11 11 ... 82 61 100 100 \n", "Tepehua 43 100 ... 96 100 74 100 \n", "Thangmi 30 12 ... 90 75 40 0 \n", "Turkana 30 25 ... 100 100 10 25 \n", "Wardaman 56 88 ... 78 100 52 100 \n", "Yimas 100 100 ... 96 100 87 61 \n", "\n", "Discharger All60/100 All50/100 All60 \n", "Local False True False True False True \n", "Language \n", "Ainu 66 75 31 0 0 0 \n", "Aleut 58 83 44 33 31 17 \n", "Ayacucho 17 50 17 50 0 50 \n", "Bella Coola 15 62 15 62 5 0 \n", "Chuckchi 26 75 26 75 16 50 \n", "Darai 0 0 0 0 0 0 \n", "Fox 44 100 0 100 0 0 \n", "Hixkaryana 73 67 73 67 36 0 \n", "Jaqaru 14 33 14 33 14 33 \n", "Jumjum 43 62 43 62 18 25 \n", "Karuk 70 50 35 25 10 0 \n", "Ket 24 0 12 0 12 0 \n", "Kunama 53 33 49 22 47 0 \n", "Lakota 0 25 0 25 0 0 \n", "Maricopa 25 50 0 0 0 0 \n", "Maung 23 0 10 0 21 0 \n", "Mordvin 50 25 50 25 40 12 \n", "Nocte 60 25 60 25 30 25 \n", "Reyesano 0 0 0 0 0 0 \n", "Sahu 54 62 38 62 0 0 \n", "Siuslawan 0 0 0 0 0 0 \n", "Tepehua 61 100 43 100 0 75 \n", "Thangmi 30 12 20 12 0 12 \n", "Turkana 20 0 20 0 20 0 \n", "Wardaman 22 50 15 25 4 0 \n", "Yimas 76 89 67 50 7 33 \n", "\n", "[26 rows x 22 columns]" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.round().astype(int)[disc]" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(xf.sort_values(by=('Intrans (norm)', True), kind='mergesort')['Intrans']\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(xf.sort_values(by=('AllPerfect (norm)', True), kind='mergesort')['AllPerfect']\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(xf.sort_values(by=('All60/100 (norm)', True), kind='mergesort')['All60/100']\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(xf.sort_values(by=('All50/100 (norm)', True), kind='mergesort')['All50/100']\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(xf.sort_values(by=('All60 (norm)', True), kind='mergesort')['All60']\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DischargerIntransFreePerfectFree60/100Free50/100Free60AllPerfectAllIntransPerfectAll60/100All50/100All60
Discharger
Intrans1.0000000.3856220.4504960.4222370.4098080.1003720.6752810.1810190.2115170.172892
FreePerfect0.3856221.0000000.5358410.3476790.3501590.6709710.4120360.3029660.2563390.217824
Free60/1000.4504960.5358411.0000000.6702340.6248520.3069960.3480540.4078940.3260370.234178
Free50/1000.4222370.3476790.6702341.0000000.8401660.2081510.2422040.5229190.6380480.428336
Free600.4098080.3501590.6248520.8401661.0000000.1630240.2513070.2498790.3202720.457837
AllPerfect0.1003720.6709710.3069960.2081510.1630241.0000000.4042200.4468340.3395130.183914
AllIntransPerfect0.6752810.4120360.3480540.2422040.2513070.4042201.0000000.1955400.2221300.119517
All60/1000.1810190.3029660.4078940.5229190.2498790.4468340.1955401.0000000.8202710.425964
All50/1000.2115170.2563390.3260370.6380480.3202720.3395130.2221300.8202711.0000000.573646
All600.1728920.2178240.2341780.4283360.4578370.1839140.1195170.4259640.5736461.000000
\n", "
" ], "text/plain": [ "Discharger Intrans FreePerfect Free60/100 Free50/100 Free60 \\\n", "Discharger \n", "Intrans 1.000000 0.385622 0.450496 0.422237 0.409808 \n", "FreePerfect 0.385622 1.000000 0.535841 0.347679 0.350159 \n", "Free60/100 0.450496 0.535841 1.000000 0.670234 0.624852 \n", "Free50/100 0.422237 0.347679 0.670234 1.000000 0.840166 \n", "Free60 0.409808 0.350159 0.624852 0.840166 1.000000 \n", "AllPerfect 0.100372 0.670971 0.306996 0.208151 0.163024 \n", "AllIntransPerfect 0.675281 0.412036 0.348054 0.242204 0.251307 \n", "All60/100 0.181019 0.302966 0.407894 0.522919 0.249879 \n", "All50/100 0.211517 0.256339 0.326037 0.638048 0.320272 \n", "All60 0.172892 0.217824 0.234178 0.428336 0.457837 \n", "\n", "Discharger AllPerfect AllIntransPerfect All60/100 All50/100 \\\n", "Discharger \n", "Intrans 0.100372 0.675281 0.181019 0.211517 \n", "FreePerfect 0.670971 0.412036 0.302966 0.256339 \n", "Free60/100 0.306996 0.348054 0.407894 0.326037 \n", "Free50/100 0.208151 0.242204 0.522919 0.638048 \n", "Free60 0.163024 0.251307 0.249879 0.320272 \n", "AllPerfect 1.000000 0.404220 0.446834 0.339513 \n", "AllIntransPerfect 0.404220 1.000000 0.195540 0.222130 \n", "All60/100 0.446834 0.195540 1.000000 0.820271 \n", "All50/100 0.339513 0.222130 0.820271 1.000000 \n", "All60 0.183914 0.119517 0.425964 0.573646 \n", "\n", "Discharger All60 \n", "Discharger \n", "Intrans 0.172892 \n", "FreePerfect 0.217824 \n", "Free60/100 0.234178 \n", "Free50/100 0.428336 \n", "Free60 0.457837 \n", "AllPerfect 0.183914 \n", "AllIntransPerfect 0.119517 \n", "All60/100 0.425964 \n", "All50/100 0.573646 \n", "All60 1.000000 " ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.stack('Local', future_stack=True).corr().loc[disc[1:], disc[1:]]" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Local & Ratiorp
Discharger
NoneNaNNaN
Intrans0.1236100.382633
FreePerfect-0.0668920.637520
Free60/1000.1719010.223012
Free50/1000.0828540.559254
Free600.0759470.592564
AllPerfect-0.1374210.331306
AllIntransPerfect-0.0065030.963507
All60/1000.1391230.325293
All50/1000.1240820.380807
All600.0248680.861083
\n", "
" ], "text/plain": [ "Local & Ratio r p\n", "Discharger \n", "None NaN NaN\n", "Intrans 0.123610 0.382633\n", "FreePerfect -0.066892 0.637520\n", "Free60/100 0.171901 0.223012\n", "Free50/100 0.082854 0.559254\n", "Free60 0.075947 0.592564\n", "AllPerfect -0.137421 0.331306\n", "AllIntransPerfect -0.006503 0.963507\n", "All60/100 0.139123 0.325293\n", "All50/100 0.124082 0.380807\n", "All60 0.024868 0.861083" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(df.stack(['Local', 'Discharger'], future_stack=True)\n", " .to_frame('Ratio')\n", " .reset_index('Local')\n", " .groupby(level='Discharger')\n", " .apply(lambda x: pearsonr(x, 'Local', 'Ratio'))\n", " .loc[disc])" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Local & Ratiorp
Discharger
NoneNaNNaN
Intrans0.1407720.329511
FreePerfect-0.0668920.637520
Free60/1000.1934800.178217
Free50/1000.0956860.517669
Free600.0946290.541198
AllPerfect-0.1374210.331306
AllIntransPerfect-0.0069270.961920
All60/1000.1702030.258107
All50/1000.1527710.322164
All600.0339800.844030
\n", "
" ], "text/plain": [ "Local & Ratio r p\n", "Discharger \n", "None NaN NaN\n", "Intrans 0.140772 0.329511\n", "FreePerfect -0.066892 0.637520\n", "Free60/100 0.193480 0.178217\n", "Free50/100 0.095686 0.517669\n", "Free60 0.094629 0.541198\n", "AllPerfect -0.137421 0.331306\n", "AllIntransPerfect -0.006927 0.961920\n", "All60/100 0.170203 0.258107\n", "All50/100 0.152771 0.322164\n", "All60 0.033980 0.844030" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(df.where(df.T.groupby(level='Discharger').any().T)\n", " .stack(['Local', 'Discharger'], future_stack=True)\n", " .to_frame('Ratio')\n", " .reset_index('Local')\n", " .groupby(level='Discharger')\n", " .apply(lambda x: pearsonr(x, 'Local', 'Ratio'))\n", " .loc[disc])" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DischargerNoneIntransFreePerfectFree60/100Free50/100...AllPerfectAllIntransPerfectAll60/100All50/100All60
LocalFalseTrueFalseTrueFalseTrueFalseTrueFalseTrue...FalseTrueFalseTrueFalseTrueFalseTrueFalseTrue
Language
Ainu0.50.50.570.430.530.470.470.531.000.00...0.530.471.000.000.470.531.000.00NaNNaN
Aleut0.50.51.000.000.460.540.410.590.570.43...0.460.541.000.000.410.590.570.430.650.35
Ayacucho0.50.50.290.710.440.560.330.670.330.67...0.480.520.180.820.250.750.250.750.001.00
Bella Coola0.50.50.500.500.470.530.190.810.190.81...0.470.530.530.470.190.810.190.811.000.00
Chuckchi0.50.50.670.330.500.500.260.740.260.74...0.500.500.670.330.260.740.260.740.240.76
Darai0.50.50.740.260.760.240.760.240.760.24...0.740.260.740.26NaNNaNNaNNaNNaNNaN
Fox0.50.50.500.500.500.500.480.520.430.57...0.500.500.500.500.300.700.001.00NaNNaN
Hixkaryana0.50.50.520.480.480.520.520.480.520.48...0.480.520.520.480.520.480.520.481.000.00
Jaqaru0.50.50.300.700.460.540.360.640.360.64...0.360.641.000.000.300.700.300.700.300.70
Jumjum0.50.50.390.610.460.540.440.560.440.56...0.460.540.360.640.410.590.410.590.420.58
Karuk0.50.50.380.620.520.480.550.450.470.53...0.520.480.390.610.580.420.580.421.000.00
Ket0.50.50.470.530.480.520.380.620.380.62...1.000.001.000.001.000.001.000.001.000.00
Kunama0.50.50.480.520.480.520.520.480.690.31...0.560.440.490.510.610.390.690.311.000.00
Lakota0.50.50.430.570.470.530.001.000.001.00...0.470.530.260.740.001.000.001.00NaNNaN
Maricopa0.50.50.200.800.500.500.330.67NaNNaN...0.500.500.200.800.330.67NaNNaNNaNNaN
Maung0.50.50.450.550.500.500.520.480.840.16...0.500.500.450.551.000.001.000.001.000.00
Mordvin0.50.50.530.470.500.500.690.310.690.31...0.530.470.510.490.670.330.670.330.760.24
Nocte0.50.50.380.620.640.360.710.290.710.29...0.640.360.440.560.710.290.710.290.550.45
Reyesano0.50.5NaNNaN1.000.00NaNNaNNaNNaN...1.000.00NaNNaNNaNNaNNaNNaNNaNNaN
Sahu0.50.50.520.480.470.530.460.540.380.62...0.470.531.000.000.460.540.380.62NaNNaN
Siuslawan0.50.50.500.500.480.520.480.520.490.51...0.570.430.500.50NaNNaNNaNNaNNaNNaN
Tepehua0.50.50.420.570.490.510.380.620.300.70...0.490.510.420.570.380.620.300.700.001.00
Thangmi0.50.50.670.330.550.450.710.290.710.29...0.550.451.000.000.710.290.620.380.001.00
Turkana0.50.50.290.710.500.500.410.590.550.45...0.500.500.290.711.000.001.000.001.000.00
Wardaman0.50.50.430.570.460.540.430.570.390.61...0.440.560.340.660.310.690.370.631.000.00
Yimas0.50.50.500.500.500.500.500.500.500.50...0.490.510.590.410.460.540.570.430.170.83
\n", "

26 rows × 22 columns

\n", "
" ], "text/plain": [ "Discharger None Intrans FreePerfect Free60/100 \\\n", "Local False True False True False True False True \n", "Language \n", "Ainu 0.5 0.5 0.57 0.43 0.53 0.47 0.47 0.53 \n", "Aleut 0.5 0.5 1.00 0.00 0.46 0.54 0.41 0.59 \n", "Ayacucho 0.5 0.5 0.29 0.71 0.44 0.56 0.33 0.67 \n", "Bella Coola 0.5 0.5 0.50 0.50 0.47 0.53 0.19 0.81 \n", "Chuckchi 0.5 0.5 0.67 0.33 0.50 0.50 0.26 0.74 \n", "Darai 0.5 0.5 0.74 0.26 0.76 0.24 0.76 0.24 \n", "Fox 0.5 0.5 0.50 0.50 0.50 0.50 0.48 0.52 \n", "Hixkaryana 0.5 0.5 0.52 0.48 0.48 0.52 0.52 0.48 \n", "Jaqaru 0.5 0.5 0.30 0.70 0.46 0.54 0.36 0.64 \n", "Jumjum 0.5 0.5 0.39 0.61 0.46 0.54 0.44 0.56 \n", "Karuk 0.5 0.5 0.38 0.62 0.52 0.48 0.55 0.45 \n", "Ket 0.5 0.5 0.47 0.53 0.48 0.52 0.38 0.62 \n", "Kunama 0.5 0.5 0.48 0.52 0.48 0.52 0.52 0.48 \n", "Lakota 0.5 0.5 0.43 0.57 0.47 0.53 0.00 1.00 \n", "Maricopa 0.5 0.5 0.20 0.80 0.50 0.50 0.33 0.67 \n", "Maung 0.5 0.5 0.45 0.55 0.50 0.50 0.52 0.48 \n", "Mordvin 0.5 0.5 0.53 0.47 0.50 0.50 0.69 0.31 \n", "Nocte 0.5 0.5 0.38 0.62 0.64 0.36 0.71 0.29 \n", "Reyesano 0.5 0.5 NaN NaN 1.00 0.00 NaN NaN \n", "Sahu 0.5 0.5 0.52 0.48 0.47 0.53 0.46 0.54 \n", "Siuslawan 0.5 0.5 0.50 0.50 0.48 0.52 0.48 0.52 \n", "Tepehua 0.5 0.5 0.42 0.57 0.49 0.51 0.38 0.62 \n", "Thangmi 0.5 0.5 0.67 0.33 0.55 0.45 0.71 0.29 \n", "Turkana 0.5 0.5 0.29 0.71 0.50 0.50 0.41 0.59 \n", "Wardaman 0.5 0.5 0.43 0.57 0.46 0.54 0.43 0.57 \n", "Yimas 0.5 0.5 0.50 0.50 0.50 0.50 0.50 0.50 \n", "\n", "Discharger Free50/100 ... AllPerfect AllIntransPerfect \\\n", "Local False True ... False True False True \n", "Language ... \n", "Ainu 1.00 0.00 ... 0.53 0.47 1.00 0.00 \n", "Aleut 0.57 0.43 ... 0.46 0.54 1.00 0.00 \n", "Ayacucho 0.33 0.67 ... 0.48 0.52 0.18 0.82 \n", "Bella Coola 0.19 0.81 ... 0.47 0.53 0.53 0.47 \n", "Chuckchi 0.26 0.74 ... 0.50 0.50 0.67 0.33 \n", "Darai 0.76 0.24 ... 0.74 0.26 0.74 0.26 \n", "Fox 0.43 0.57 ... 0.50 0.50 0.50 0.50 \n", "Hixkaryana 0.52 0.48 ... 0.48 0.52 0.52 0.48 \n", "Jaqaru 0.36 0.64 ... 0.36 0.64 1.00 0.00 \n", "Jumjum 0.44 0.56 ... 0.46 0.54 0.36 0.64 \n", "Karuk 0.47 0.53 ... 0.52 0.48 0.39 0.61 \n", "Ket 0.38 0.62 ... 1.00 0.00 1.00 0.00 \n", "Kunama 0.69 0.31 ... 0.56 0.44 0.49 0.51 \n", "Lakota 0.00 1.00 ... 0.47 0.53 0.26 0.74 \n", "Maricopa NaN NaN ... 0.50 0.50 0.20 0.80 \n", "Maung 0.84 0.16 ... 0.50 0.50 0.45 0.55 \n", "Mordvin 0.69 0.31 ... 0.53 0.47 0.51 0.49 \n", "Nocte 0.71 0.29 ... 0.64 0.36 0.44 0.56 \n", "Reyesano NaN NaN ... 1.00 0.00 NaN NaN \n", "Sahu 0.38 0.62 ... 0.47 0.53 1.00 0.00 \n", "Siuslawan 0.49 0.51 ... 0.57 0.43 0.50 0.50 \n", "Tepehua 0.30 0.70 ... 0.49 0.51 0.42 0.57 \n", "Thangmi 0.71 0.29 ... 0.55 0.45 1.00 0.00 \n", "Turkana 0.55 0.45 ... 0.50 0.50 0.29 0.71 \n", "Wardaman 0.39 0.61 ... 0.44 0.56 0.34 0.66 \n", "Yimas 0.50 0.50 ... 0.49 0.51 0.59 0.41 \n", "\n", "Discharger All60/100 All50/100 All60 \n", "Local False True False True False True \n", "Language \n", "Ainu 0.47 0.53 1.00 0.00 NaN NaN \n", "Aleut 0.41 0.59 0.57 0.43 0.65 0.35 \n", "Ayacucho 0.25 0.75 0.25 0.75 0.00 1.00 \n", "Bella Coola 0.19 0.81 0.19 0.81 1.00 0.00 \n", "Chuckchi 0.26 0.74 0.26 0.74 0.24 0.76 \n", "Darai NaN NaN NaN NaN NaN NaN \n", "Fox 0.30 0.70 0.00 1.00 NaN NaN \n", "Hixkaryana 0.52 0.48 0.52 0.48 1.00 0.00 \n", "Jaqaru 0.30 0.70 0.30 0.70 0.30 0.70 \n", "Jumjum 0.41 0.59 0.41 0.59 0.42 0.58 \n", "Karuk 0.58 0.42 0.58 0.42 1.00 0.00 \n", "Ket 1.00 0.00 1.00 0.00 1.00 0.00 \n", "Kunama 0.61 0.39 0.69 0.31 1.00 0.00 \n", "Lakota 0.00 1.00 0.00 1.00 NaN NaN \n", "Maricopa 0.33 0.67 NaN NaN NaN NaN \n", "Maung 1.00 0.00 1.00 0.00 1.00 0.00 \n", "Mordvin 0.67 0.33 0.67 0.33 0.76 0.24 \n", "Nocte 0.71 0.29 0.71 0.29 0.55 0.45 \n", "Reyesano NaN NaN NaN NaN NaN NaN \n", "Sahu 0.46 0.54 0.38 0.62 NaN NaN \n", "Siuslawan NaN NaN NaN NaN NaN NaN \n", "Tepehua 0.38 0.62 0.30 0.70 0.00 1.00 \n", "Thangmi 0.71 0.29 0.62 0.38 0.00 1.00 \n", "Turkana 1.00 0.00 1.00 0.00 1.00 0.00 \n", "Wardaman 0.31 0.69 0.37 0.63 1.00 0.00 \n", "Yimas 0.46 0.54 0.57 0.43 0.17 0.83 \n", "\n", "[26 rows x 22 columns]" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "nf.round(2)[disc]" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Local & Ratiorp
Discharger  
Nonenannan
Intrans0.0940430.515945
FreePerfect-0.2014860.152042
Free60/1000.2817790.047429
Free50/1000.0094900.948958
Free600.1223060.428998
AllPerfect-0.3026170.029216
AllIntransPerfect-0.2686900.059197
All60/1000.0288650.848976
All50/100-0.0584140.706442
All60-0.2800940.098013
\n" ], "text/plain": [ "" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(nf.stack(['Local', 'Discharger'], future_stack=True)\n", " .to_frame('Ratio')\n", " .reset_index('Local')\n", " .groupby(level='Discharger')\n", " .apply(lambda x: pearsonr(x, 'Local', 'Ratio'))\n", " .loc[disc]\n", " .style.map(lambda x: x < .05 and 'background-color: lime' or '', subset=['p'])\n", " .set_uuid('16'))" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(nf['AllPerfect'].sort_values(by=True, kind='mergesort')\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(nf['All60/100'].sort_values(by=True, kind='mergesort')\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(nf['All50/100'].sort_values(by=True, kind='mergesort')\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(nf['All60'].sort_values(by=True, kind='mergesort')\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Direct" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "dff = (tb[tb['Trans']]\n", " .pivot_table('Form', 'Language', ['Discharger', 'Direct'],\n", " aggfunc='count', fill_value=0))\n", "\n", "dff['All50/100'].plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray);" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DischargerNoneIntransFreePerfectFree60/100Free50/100...AllPerfectAllIntransPerfectAll60/100All50/100All60
DirectFalseTrueFalseTrueFalseTrueFalseTrueFalseTrue...FalseTrueFalseTrueFalseTrueFalseTrueFalseTrue
Language
Ainu0.50.50.620.380.400.600.330.670.001.00...0.400.600.410.590.330.670.001.00NaNNaN
Aleut0.50.50.001.000.550.450.360.640.290.71...0.550.450.001.000.360.640.290.710.001.00
Ayacucho0.50.50.820.180.470.530.660.340.660.34...0.450.550.660.340.560.440.560.440.001.00
Bella Coola0.50.50.550.450.530.470.690.310.690.31...0.530.470.550.450.690.310.690.311.000.00
Chuckchi0.50.50.390.610.500.500.350.650.350.65...0.500.500.390.610.350.650.350.650.200.80
Darai0.50.50.310.690.400.600.400.600.400.60...0.310.690.310.69NaNNaNNaNNaNNaNNaN
Fox0.50.50.500.500.500.500.470.530.490.51...0.500.500.500.500.500.500.550.45NaNNaN
Hixkaryana0.50.50.730.270.530.470.550.450.550.45...0.530.470.730.270.550.450.550.450.380.62
Jaqaru0.50.50.830.170.560.440.710.290.710.29...0.430.571.000.000.001.000.001.000.001.00
Jumjum0.50.50.670.330.560.440.590.410.590.41...0.560.440.640.360.470.530.470.530.330.67
Karuk0.50.50.840.160.450.550.470.530.440.56...0.450.550.820.180.400.600.001.000.001.00
Ket0.50.50.540.460.530.470.450.550.450.55...0.580.420.001.000.360.640.001.000.600.40
Kunama0.50.50.580.420.520.480.650.350.500.50...0.410.590.900.100.450.550.500.500.390.61
Lakota0.50.50.470.530.530.470.001.000.001.00...0.530.470.550.450.001.000.001.00NaNNaN
Maricopa0.50.50.670.330.500.500.500.50NaNNaN...0.500.500.670.330.500.50NaNNaNNaNNaN
Maung0.50.50.560.440.500.500.480.520.390.61...0.500.500.560.440.001.000.001.000.001.00
Mordvin0.50.50.520.480.500.500.290.710.290.71...0.320.680.510.490.210.790.210.790.280.72
Nocte0.50.50.670.330.620.380.500.500.500.50...0.620.381.000.000.500.500.500.500.570.43
Reyesano0.50.5NaNNaN0.570.43NaNNaNNaNNaN...0.570.43NaNNaNNaNNaNNaNNaNNaNNaN
Sahu0.50.50.500.500.240.760.250.750.330.67...0.240.760.001.000.250.750.330.67NaNNaN
Siuslawan0.50.50.500.500.470.530.470.530.900.10...0.400.600.500.50NaNNaNNaNNaNNaNNaN
Tepehua0.50.50.610.390.520.480.590.410.710.29...0.520.480.610.390.590.410.710.290.550.45
Thangmi0.50.50.001.000.570.430.500.500.500.50...0.570.430.001.000.500.500.670.331.000.00
Turkana0.50.50.570.430.500.500.620.380.310.69...0.500.500.570.430.001.000.001.000.001.00
Wardaman0.50.50.610.390.490.510.510.490.450.55...0.530.470.760.240.540.460.370.631.000.00
Yimas0.50.50.500.500.500.500.500.500.500.50...0.480.520.430.570.470.530.370.630.730.27
\n", "

26 rows × 22 columns

\n", "
" ], "text/plain": [ "Discharger None Intrans FreePerfect Free60/100 \\\n", "Direct False True False True False True False True \n", "Language \n", "Ainu 0.5 0.5 0.62 0.38 0.40 0.60 0.33 0.67 \n", "Aleut 0.5 0.5 0.00 1.00 0.55 0.45 0.36 0.64 \n", "Ayacucho 0.5 0.5 0.82 0.18 0.47 0.53 0.66 0.34 \n", "Bella Coola 0.5 0.5 0.55 0.45 0.53 0.47 0.69 0.31 \n", "Chuckchi 0.5 0.5 0.39 0.61 0.50 0.50 0.35 0.65 \n", "Darai 0.5 0.5 0.31 0.69 0.40 0.60 0.40 0.60 \n", "Fox 0.5 0.5 0.50 0.50 0.50 0.50 0.47 0.53 \n", "Hixkaryana 0.5 0.5 0.73 0.27 0.53 0.47 0.55 0.45 \n", "Jaqaru 0.5 0.5 0.83 0.17 0.56 0.44 0.71 0.29 \n", "Jumjum 0.5 0.5 0.67 0.33 0.56 0.44 0.59 0.41 \n", "Karuk 0.5 0.5 0.84 0.16 0.45 0.55 0.47 0.53 \n", "Ket 0.5 0.5 0.54 0.46 0.53 0.47 0.45 0.55 \n", "Kunama 0.5 0.5 0.58 0.42 0.52 0.48 0.65 0.35 \n", "Lakota 0.5 0.5 0.47 0.53 0.53 0.47 0.00 1.00 \n", "Maricopa 0.5 0.5 0.67 0.33 0.50 0.50 0.50 0.50 \n", "Maung 0.5 0.5 0.56 0.44 0.50 0.50 0.48 0.52 \n", "Mordvin 0.5 0.5 0.52 0.48 0.50 0.50 0.29 0.71 \n", "Nocte 0.5 0.5 0.67 0.33 0.62 0.38 0.50 0.50 \n", "Reyesano 0.5 0.5 NaN NaN 0.57 0.43 NaN NaN \n", "Sahu 0.5 0.5 0.50 0.50 0.24 0.76 0.25 0.75 \n", "Siuslawan 0.5 0.5 0.50 0.50 0.47 0.53 0.47 0.53 \n", "Tepehua 0.5 0.5 0.61 0.39 0.52 0.48 0.59 0.41 \n", "Thangmi 0.5 0.5 0.00 1.00 0.57 0.43 0.50 0.50 \n", "Turkana 0.5 0.5 0.57 0.43 0.50 0.50 0.62 0.38 \n", "Wardaman 0.5 0.5 0.61 0.39 0.49 0.51 0.51 0.49 \n", "Yimas 0.5 0.5 0.50 0.50 0.50 0.50 0.50 0.50 \n", "\n", "Discharger Free50/100 ... AllPerfect AllIntransPerfect \\\n", "Direct False True ... False True False True \n", "Language ... \n", "Ainu 0.00 1.00 ... 0.40 0.60 0.41 0.59 \n", "Aleut 0.29 0.71 ... 0.55 0.45 0.00 1.00 \n", "Ayacucho 0.66 0.34 ... 0.45 0.55 0.66 0.34 \n", "Bella Coola 0.69 0.31 ... 0.53 0.47 0.55 0.45 \n", "Chuckchi 0.35 0.65 ... 0.50 0.50 0.39 0.61 \n", "Darai 0.40 0.60 ... 0.31 0.69 0.31 0.69 \n", "Fox 0.49 0.51 ... 0.50 0.50 0.50 0.50 \n", "Hixkaryana 0.55 0.45 ... 0.53 0.47 0.73 0.27 \n", "Jaqaru 0.71 0.29 ... 0.43 0.57 1.00 0.00 \n", "Jumjum 0.59 0.41 ... 0.56 0.44 0.64 0.36 \n", "Karuk 0.44 0.56 ... 0.45 0.55 0.82 0.18 \n", "Ket 0.45 0.55 ... 0.58 0.42 0.00 1.00 \n", "Kunama 0.50 0.50 ... 0.41 0.59 0.90 0.10 \n", "Lakota 0.00 1.00 ... 0.53 0.47 0.55 0.45 \n", "Maricopa NaN NaN ... 0.50 0.50 0.67 0.33 \n", "Maung 0.39 0.61 ... 0.50 0.50 0.56 0.44 \n", "Mordvin 0.29 0.71 ... 0.32 0.68 0.51 0.49 \n", "Nocte 0.50 0.50 ... 0.62 0.38 1.00 0.00 \n", "Reyesano NaN NaN ... 0.57 0.43 NaN NaN \n", "Sahu 0.33 0.67 ... 0.24 0.76 0.00 1.00 \n", "Siuslawan 0.90 0.10 ... 0.40 0.60 0.50 0.50 \n", "Tepehua 0.71 0.29 ... 0.52 0.48 0.61 0.39 \n", "Thangmi 0.50 0.50 ... 0.57 0.43 0.00 1.00 \n", "Turkana 0.31 0.69 ... 0.50 0.50 0.57 0.43 \n", "Wardaman 0.45 0.55 ... 0.53 0.47 0.76 0.24 \n", "Yimas 0.50 0.50 ... 0.48 0.52 0.43 0.57 \n", "\n", "Discharger All60/100 All50/100 All60 \n", "Direct False True False True False True \n", "Language \n", "Ainu 0.33 0.67 0.00 1.00 NaN NaN \n", "Aleut 0.36 0.64 0.29 0.71 0.00 1.00 \n", "Ayacucho 0.56 0.44 0.56 0.44 0.00 1.00 \n", "Bella Coola 0.69 0.31 0.69 0.31 1.00 0.00 \n", "Chuckchi 0.35 0.65 0.35 0.65 0.20 0.80 \n", "Darai NaN NaN NaN NaN NaN NaN \n", "Fox 0.50 0.50 0.55 0.45 NaN NaN \n", "Hixkaryana 0.55 0.45 0.55 0.45 0.38 0.62 \n", "Jaqaru 0.00 1.00 0.00 1.00 0.00 1.00 \n", "Jumjum 0.47 0.53 0.47 0.53 0.33 0.67 \n", "Karuk 0.40 0.60 0.00 1.00 0.00 1.00 \n", "Ket 0.36 0.64 0.00 1.00 0.60 0.40 \n", "Kunama 0.45 0.55 0.50 0.50 0.39 0.61 \n", "Lakota 0.00 1.00 0.00 1.00 NaN NaN \n", "Maricopa 0.50 0.50 NaN NaN NaN NaN \n", "Maung 0.00 1.00 0.00 1.00 0.00 1.00 \n", "Mordvin 0.21 0.79 0.21 0.79 0.28 0.72 \n", "Nocte 0.50 0.50 0.50 0.50 0.57 0.43 \n", "Reyesano NaN NaN NaN NaN NaN NaN \n", "Sahu 0.25 0.75 0.33 0.67 NaN NaN \n", "Siuslawan NaN NaN NaN NaN NaN NaN \n", "Tepehua 0.59 0.41 0.71 0.29 0.55 0.45 \n", "Thangmi 0.50 0.50 0.67 0.33 1.00 0.00 \n", "Turkana 0.00 1.00 0.00 1.00 0.00 1.00 \n", "Wardaman 0.54 0.46 0.37 0.63 1.00 0.00 \n", "Yimas 0.47 0.53 0.37 0.63 0.73 0.27 \n", "\n", "[26 rows x 22 columns]" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ddf = 100 * dff / dff['None']\n", "dnf = ddf / ddf.T.groupby(level='Discharger').sum().T\n", "dnf.round(2)[disc]" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Direct & Ratiorp
Discharger  
Nonenannan
Intrans-0.2042420.154814
FreePerfect-0.0010380.994175
Free60/1000.1600600.266849
Free50/1000.2033840.165599
Free600.3066440.042913
AllPerfect0.2282500.103633
AllIntransPerfect-0.0775280.592546
All60/1000.5319550.000142
All50/1000.5747280.000045
All600.2941960.081569
\n" ], "text/plain": [ "" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(dnf.stack(['Direct', 'Discharger'], future_stack=True)\n", " .to_frame('Ratio')\n", " .reset_index('Direct')\n", " .groupby(level='Discharger')\n", " .apply(lambda x: pearsonr(x, 'Direct', 'Ratio'))\n", " .loc[disc]\n", " .style.map(lambda x: x < .05 and 'background-color: lime' or '', subset=['p'])\n", " .set_uuid('21'))" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(dnf['Intrans'].sort_values(by=True, kind='mergesort')\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(dnf['AllPerfect'].sort_values(by=True, kind='mergesort')\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(dnf['All60/100'].sort_values(by=True, kind='mergesort')\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "(dnf['All50/100'].sort_values(by=True, kind='mergesort')\n", " .plot.bar(stacked=True, figsize=(15, 5), cmap=plt.cm.gray));" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Local & form contingency" ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
DischargerNoneIntransFreePerfectFree60/100Free50/100...AllPerfectAllIntransPerfectAll60/100All50/100All60
LocalFalseTrueFalseTrueFalseTrueFalseTrueFalseTrue...FalseTrueFalseTrueFalseTrueFalseTrueFalseTrue
LanguageForm
AinuFalse0010442102208...42148102208298
True29819425619690...2561501969000
AleutFalse001818601932512...60181819325123115
True451827039182615206...39182702615206143
AyacuchoFalse0014050152152...51202204204244
True2481081989696...19746444404
Bella CoolaFalse005220173173...2063173173198
True2081561883535...188145353510
ChuckchiFalse000400142142...0004142142164
True1981941985656...198194565634
DaraiFalse0066464646...6666208208208
True208142162162162...142142000000
FoxFalse0000002080...0000180320328
True328328328308248...3283281480800
HixkaryanaFalse00144206262...201446262146
True22682206164164...2068216416480
JaqaruFalse0040103030...3053626262
True7333634343...4320111111
JumjumFalse00100406060...40120163163236
True288188248228228...24816812512552
KarukFalse00112115393...1112364136188
True20896197155115...197851447220
KetFalse0010080320320...208748648748748
True848748768528528...640100200100100
KunamaFalse00102501562914...25104514271229143018
True5718471652184212284...328124306284270
LakotaFalse0010251236236...51194236236238
True2381361870202...18744020200
MaricopaFalse0012000124168...00120124168168
True168481684400...16848440000
MaungFalse001600051337...10160748868768
True968808968917631...958808220100200
MordvinFalse0001009696...6321106106127
True208207208112112...14518710210281
NocteFalse0080248686...241248686146
True208128184122122...1848412212262
ReyesanoFalse0020848208208...48208208208208
True208001600000...16000000000
SahuFalse00172423443603...423638443603968
True968796545525365...54533052536500
SiuslawanFalse000040405116...10700571857185718
True571857185318531862...47115718000000
TepehuaFalse00601090130...106090130232
True238178228148108...22817814810806
ThangmiFalse0010622147147...22128147167207
True2081021866161...18680614101
TurkanaFalse001860060146...00186168168168
True2082220814862...20822404040
WardamanFalse00704070121...60130214236268
True278208238208157...218148644210
YimasFalse0000000000...20671121594212
True45184518451845184518...43183911341630936
\n", "

52 rows × 22 columns

\n", "
" ], "text/plain": [ "Discharger None Intrans FreePerfect Free60/100 \\\n", "Local False True False True False True False \n", "Language Form \n", "Ainu False 0 0 10 4 4 2 10 \n", " True 29 8 19 4 25 6 19 \n", "Aleut False 0 0 18 18 6 0 19 \n", " True 45 18 27 0 39 18 26 \n", "Ayacucho False 0 0 14 0 5 0 15 \n", " True 24 8 10 8 19 8 9 \n", "Bella Coola False 0 0 5 2 2 0 17 \n", " True 20 8 15 6 18 8 3 \n", "Chuckchi False 0 0 0 4 0 0 14 \n", " True 19 8 19 4 19 8 5 \n", "Darai False 0 0 6 6 4 6 4 \n", " True 20 8 14 2 16 2 16 \n", "Fox False 0 0 0 0 0 0 2 \n", " True 32 8 32 8 32 8 30 \n", "Hixkaryana False 0 0 14 4 2 0 6 \n", " True 22 6 8 2 20 6 16 \n", "Jaqaru False 0 0 4 0 1 0 3 \n", " True 7 3 3 3 6 3 4 \n", "Jumjum False 0 0 10 0 4 0 6 \n", " True 28 8 18 8 24 8 22 \n", "Karuk False 0 0 11 2 1 1 5 \n", " True 20 8 9 6 19 7 15 \n", "Ket False 0 0 10 0 8 0 32 \n", " True 84 8 74 8 76 8 52 \n", "Kunama False 0 0 10 2 5 0 15 \n", " True 57 18 47 16 52 18 42 \n", "Lakota False 0 0 10 2 5 1 23 \n", " True 23 8 13 6 18 7 0 \n", "Maricopa False 0 0 12 0 0 0 12 \n", " True 16 8 4 8 16 8 4 \n", "Maung False 0 0 16 0 0 0 5 \n", " True 96 8 80 8 96 8 91 \n", "Mordvin False 0 0 0 1 0 0 9 \n", " True 20 8 20 7 20 8 11 \n", "Nocte False 0 0 8 0 2 4 8 \n", " True 20 8 12 8 18 4 12 \n", "Reyesano False 0 0 20 8 4 8 20 \n", " True 20 8 0 0 16 0 0 \n", "Sahu False 0 0 17 2 42 3 44 \n", " True 96 8 79 6 54 5 52 \n", "Siuslawan False 0 0 0 0 4 0 4 \n", " True 57 18 57 18 53 18 53 \n", "Tepehua False 0 0 6 0 1 0 9 \n", " True 23 8 17 8 22 8 14 \n", "Thangmi False 0 0 10 6 2 2 14 \n", " True 20 8 10 2 18 6 6 \n", "Turkana False 0 0 18 6 0 0 6 \n", " True 20 8 2 2 20 8 14 \n", "Wardaman False 0 0 7 0 4 0 7 \n", " True 27 8 20 8 23 8 20 \n", "Yimas False 0 0 0 0 0 0 0 \n", " True 45 18 45 18 45 18 45 \n", "\n", "Discharger Free50/100 ... AllPerfect \\\n", "Local True False True ... False True \n", "Language Form ... \n", "Ainu False 2 20 8 ... 4 2 \n", " True 6 9 0 ... 25 6 \n", "Aleut False 3 25 12 ... 6 0 \n", " True 15 20 6 ... 39 18 \n", "Ayacucho False 2 15 2 ... 5 1 \n", " True 6 9 6 ... 19 7 \n", "Bella Coola False 3 17 3 ... 2 0 \n", " True 5 3 5 ... 18 8 \n", "Chuckchi False 2 14 2 ... 0 0 \n", " True 6 5 6 ... 19 8 \n", "Darai False 6 4 6 ... 6 6 \n", " True 2 16 2 ... 14 2 \n", "Fox False 0 8 0 ... 0 0 \n", " True 8 24 8 ... 32 8 \n", "Hixkaryana False 2 6 2 ... 2 0 \n", " True 4 16 4 ... 20 6 \n", "Jaqaru False 0 3 0 ... 3 0 \n", " True 3 4 3 ... 4 3 \n", "Jumjum False 0 6 0 ... 4 0 \n", " True 8 22 8 ... 24 8 \n", "Karuk False 3 9 3 ... 1 1 \n", " True 5 11 5 ... 19 7 \n", "Ket False 0 32 0 ... 20 8 \n", " True 8 52 8 ... 64 0 \n", "Kunama False 6 29 14 ... 25 10 \n", " True 12 28 4 ... 32 8 \n", "Lakota False 6 23 6 ... 5 1 \n", " True 2 0 2 ... 18 7 \n", "Maricopa False 4 16 8 ... 0 0 \n", " True 4 0 0 ... 16 8 \n", "Maung False 1 33 7 ... 1 0 \n", " True 7 63 1 ... 95 8 \n", "Mordvin False 6 9 6 ... 6 3 \n", " True 2 11 2 ... 14 5 \n", "Nocte False 6 8 6 ... 2 4 \n", " True 2 12 2 ... 18 4 \n", "Reyesano False 8 20 8 ... 4 8 \n", " True 0 0 0 ... 16 0 \n", "Sahu False 3 60 3 ... 42 3 \n", " True 5 36 5 ... 54 5 \n", "Siuslawan False 0 51 16 ... 10 7 \n", " True 18 6 2 ... 47 11 \n", "Tepehua False 0 13 0 ... 1 0 \n", " True 8 10 8 ... 22 8 \n", "Thangmi False 7 14 7 ... 2 2 \n", " True 1 6 1 ... 18 6 \n", "Turkana False 0 14 6 ... 0 0 \n", " True 8 6 2 ... 20 8 \n", "Wardaman False 0 12 1 ... 6 0 \n", " True 8 15 7 ... 21 8 \n", "Yimas False 0 0 0 ... 2 0 \n", " True 18 45 18 ... 43 18 \n", "\n", "Discharger AllIntransPerfect All60/100 All50/100 \\\n", "Local False True False True False True \n", "Language Form \n", "Ainu False 14 8 10 2 20 8 \n", " True 15 0 19 6 9 0 \n", "Aleut False 18 18 19 3 25 12 \n", " True 27 0 26 15 20 6 \n", "Ayacucho False 20 2 20 4 20 4 \n", " True 4 6 4 4 4 4 \n", "Bella Coola False 6 3 17 3 17 3 \n", " True 14 5 3 5 3 5 \n", "Chuckchi False 0 4 14 2 14 2 \n", " True 19 4 5 6 5 6 \n", "Darai False 6 6 20 8 20 8 \n", " True 14 2 0 0 0 0 \n", "Fox False 0 0 18 0 32 0 \n", " True 32 8 14 8 0 8 \n", "Hixkaryana False 14 4 6 2 6 2 \n", " True 8 2 16 4 16 4 \n", "Jaqaru False 5 3 6 2 6 2 \n", " True 2 0 1 1 1 1 \n", "Jumjum False 12 0 16 3 16 3 \n", " True 16 8 12 5 12 5 \n", "Karuk False 12 3 6 4 13 6 \n", " True 8 5 14 4 7 2 \n", "Ket False 74 8 64 8 74 8 \n", " True 10 0 20 0 10 0 \n", "Kunama False 45 14 27 12 29 14 \n", " True 12 4 30 6 28 4 \n", "Lakota False 19 4 23 6 23 6 \n", " True 4 4 0 2 0 2 \n", "Maricopa False 12 0 12 4 16 8 \n", " True 4 8 4 4 0 0 \n", "Maung False 16 0 74 8 86 8 \n", " True 80 8 22 0 10 0 \n", "Mordvin False 2 1 10 6 10 6 \n", " True 18 7 10 2 10 2 \n", "Nocte False 12 4 8 6 8 6 \n", " True 8 4 12 2 12 2 \n", "Reyesano False 20 8 20 8 20 8 \n", " True 0 0 0 0 0 0 \n", "Sahu False 63 8 44 3 60 3 \n", " True 33 0 52 5 36 5 \n", "Siuslawan False 0 0 57 18 57 18 \n", " True 57 18 0 0 0 0 \n", "Tepehua False 6 0 9 0 13 0 \n", " True 17 8 14 8 10 8 \n", "Thangmi False 12 8 14 7 16 7 \n", " True 8 0 6 1 4 1 \n", "Turkana False 18 6 16 8 16 8 \n", " True 2 2 4 0 4 0 \n", "Wardaman False 13 0 21 4 23 6 \n", " True 14 8 6 4 4 2 \n", "Yimas False 6 7 11 2 15 9 \n", " True 39 11 34 16 30 9 \n", "\n", "Discharger All60 \n", "Local False True \n", "Language Form \n", "Ainu False 29 8 \n", " True 0 0 \n", "Aleut False 31 15 \n", " True 14 3 \n", "Ayacucho False 24 4 \n", " True 0 4 \n", "Bella Coola False 19 8 \n", " True 1 0 \n", "Chuckchi False 16 4 \n", " True 3 4 \n", "Darai False 20 8 \n", " True 0 0 \n", "Fox False 32 8 \n", " True 0 0 \n", "Hixkaryana False 14 6 \n", " True 8 0 \n", "Jaqaru False 6 2 \n", " True 1 1 \n", "Jumjum False 23 6 \n", " True 5 2 \n", "Karuk False 18 8 \n", " True 2 0 \n", "Ket False 74 8 \n", " True 10 0 \n", "Kunama False 30 18 \n", " True 27 0 \n", "Lakota False 23 8 \n", " True 0 0 \n", "Maricopa False 16 8 \n", " True 0 0 \n", "Maung False 76 8 \n", " True 20 0 \n", "Mordvin False 12 7 \n", " True 8 1 \n", "Nocte False 14 6 \n", " True 6 2 \n", "Reyesano False 20 8 \n", " True 0 0 \n", "Sahu False 96 8 \n", " True 0 0 \n", "Siuslawan False 57 18 \n", " True 0 0 \n", "Tepehua False 23 2 \n", " True 0 6 \n", "Thangmi False 20 7 \n", " True 0 1 \n", "Turkana False 16 8 \n", " True 4 0 \n", "Wardaman False 26 8 \n", " True 1 0 \n", "Yimas False 42 12 \n", " True 3 6 \n", "\n", "[52 rows x 22 columns]" ] }, "execution_count": 31, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ctl = (tb[tb['Trans']]\n", " .set_index(['Language', 'Discharger', 'Cell', 'Local'])['Form'].notnull()\n", " .unstack('Discharger')\n", " .rename(columns={'None': 'Base'})\n", " .query('Base == 1')\n", " .rename(columns={'Base': 'None'})\n", " .stack('Discharger')\n", " .to_frame('Form')\n", " .reset_index()\n", " .pivot_table('Cell', ['Language', 'Form'], ['Discharger', 'Local'],\n", " aggfunc=len, fill_value=0))\n", "\n", "ctl[disc]" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
oddsp
iDischargerLanguage
1IntransAleut0.0000006.797740e-06
Ayacuchoinf4.445680e-03
Chuckchi0.0000003.988604e-03
Darai0.1428574.405160e-02
Maricopainf1.346076e-03
2FreePerfectDarai0.0833331.104692e-02
Nocte0.1111113.835006e-02
Reyesano0.0000001.592610e-04
3Free60/100Bella Coola9.4444442.230330e-02
Chuckchi8.4000003.322185e-02
Darai0.0833331.104692e-02
Ketinf4.700267e-02
4Free50/100Bella Coola9.4444442.230330e-02
Chuckchi8.4000003.322185e-02
Darai0.0833331.104692e-02
Ketinf4.700267e-02
Maung0.0748304.930952e-03
Tepehuainf9.625510e-03
5Free60Ayacucho11.4000009.898463e-03
Bella Coola9.4444442.230330e-02
Darai0.0833331.104692e-02
Ketinf1.902499e-02
Kunama0.0000001.205725e-04
Maung0.0933333.636695e-03
Siuslawaninf1.208441e-02
Tepehuainf3.802896e-05
6AllPerfectDarai0.1428574.405160e-02
Ket0.0000003.339144e-05
Nocte0.1111113.835006e-02
Reyesano0.0000001.592610e-04
7AllIntransPerfectAinu0.0000001.211601e-02
Aleut0.0000006.797740e-06
Ayacucho15.0000004.867231e-03
Chuckchi0.0000003.988604e-03
Darai0.1428574.405160e-02
Jumjuminf3.337695e-02
Maricopainf1.346076e-03
Wardamaninf1.524523e-02
Yimas0.2417583.759624e-02
8All60/100Bella Coola9.4444442.230330e-02
Chuckchi8.4000003.322185e-02
Foxinf4.726994e-03
9All50/100Bella Coola9.4444442.230330e-02
Chuckchi8.4000003.322185e-02
Foxinf1.300311e-08
Tepehuainf9.625510e-03
10All60Ayacuchoinf1.946607e-03
Kunama0.0000001.205725e-04
Tepehuainf3.802896e-05
Yimas7.0000001.254650e-02
\n", "
" ], "text/plain": [ " odds p\n", "i Discharger Language \n", "1 Intrans Aleut 0.000000 6.797740e-06\n", " Ayacucho inf 4.445680e-03\n", " Chuckchi 0.000000 3.988604e-03\n", " Darai 0.142857 4.405160e-02\n", " Maricopa inf 1.346076e-03\n", "2 FreePerfect Darai 0.083333 1.104692e-02\n", " Nocte 0.111111 3.835006e-02\n", " Reyesano 0.000000 1.592610e-04\n", "3 Free60/100 Bella Coola 9.444444 2.230330e-02\n", " Chuckchi 8.400000 3.322185e-02\n", " Darai 0.083333 1.104692e-02\n", " Ket inf 4.700267e-02\n", "4 Free50/100 Bella Coola 9.444444 2.230330e-02\n", " Chuckchi 8.400000 3.322185e-02\n", " Darai 0.083333 1.104692e-02\n", " Ket inf 4.700267e-02\n", " Maung 0.074830 4.930952e-03\n", " Tepehua inf 9.625510e-03\n", "5 Free60 Ayacucho 11.400000 9.898463e-03\n", " Bella Coola 9.444444 2.230330e-02\n", " Darai 0.083333 1.104692e-02\n", " Ket inf 1.902499e-02\n", " Kunama 0.000000 1.205725e-04\n", " Maung 0.093333 3.636695e-03\n", " Siuslawan inf 1.208441e-02\n", " Tepehua inf 3.802896e-05\n", "6 AllPerfect Darai 0.142857 4.405160e-02\n", " Ket 0.000000 3.339144e-05\n", " Nocte 0.111111 3.835006e-02\n", " Reyesano 0.000000 1.592610e-04\n", "7 AllIntransPerfect Ainu 0.000000 1.211601e-02\n", " Aleut 0.000000 6.797740e-06\n", " Ayacucho 15.000000 4.867231e-03\n", " Chuckchi 0.000000 3.988604e-03\n", " Darai 0.142857 4.405160e-02\n", " Jumjum inf 3.337695e-02\n", " Maricopa inf 1.346076e-03\n", " Wardaman inf 1.524523e-02\n", " Yimas 0.241758 3.759624e-02\n", "8 All60/100 Bella Coola 9.444444 2.230330e-02\n", " Chuckchi 8.400000 3.322185e-02\n", " Fox inf 4.726994e-03\n", "9 All50/100 Bella Coola 9.444444 2.230330e-02\n", " Chuckchi 8.400000 3.322185e-02\n", " Fox inf 1.300311e-08\n", " Tepehua inf 9.625510e-03\n", "10 All60 Ayacucho inf 1.946607e-03\n", " Kunama 0.000000 1.205725e-04\n", " Tepehua inf 3.802896e-05\n", " Yimas 7.000000 1.254650e-02" ] }, "execution_count": 32, "metadata": {}, "output_type": "execute_result" } ], "source": [ "(ctl.stack('Discharger', future_stack=True)\n", " .groupby(level=['Discharger', 'Language'])\n", " .apply(fisher_exact).query('p <= .05')\n", " .assign(i=lambda x: list(map(disc.tolist().index, x.index.get_level_values('Discharger'))))\n", " .reset_index()\n", " .set_index(['i', 'Discharger', 'Language'])\n", " .sort_index())" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.14.0" } }, "nbformat": 4, "nbformat_minor": 4 }