diff --git a/.ipynb_checkpoints/grad_school-checkpoint.ipynb b/.ipynb_checkpoints/grad_school-checkpoint.ipynb new file mode 100644 index 0000000..93ce608 --- /dev/null +++ b/.ipynb_checkpoints/grad_school-checkpoint.ipynb @@ -0,0 +1,585 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 66, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "from sklearn.linear_model import LogisticRegression\n", + "from sklearn.metrics import confusion_matrix\n", + "from sklearn.model_selection import train_test_split\n", + "from pandas.plotting import scatter_matrix\n", + "from itertools import combinations\n", + "from sklearn import linear_model\n", + "from sklearn.model_selection import cross_val_score\n", + "%matplotlib inline" + ] + }, + { + "cell_type": "code", + "execution_count": 67, + "metadata": {}, + "outputs": [], + "source": [ + "df = pd.read_csv('grad.csv')" + ] + }, + { + "cell_type": "code", + "execution_count": 68, + "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", + "
admitgregparank
003803.613
116603.673
218004.001
316403.194
405202.934
\n", + "
" + ], + "text/plain": [ + " admit gre gpa rank\n", + "0 0 380 3.61 3\n", + "1 1 660 3.67 3\n", + "2 1 800 4.00 1\n", + "3 1 640 3.19 4\n", + "4 0 520 2.93 4" + ] + }, + "execution_count": 68, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "y = df.admit\n", + "X = df[['gpa','gre','rank']]\n", + "df.head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 69, + "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", + "
admitgregparank
count400.000000400.000000400.000000400.00000
mean0.317500587.7000003.3899002.48500
std0.466087115.5165360.3805670.94446
min0.000000220.0000002.2600001.00000
25%0.000000520.0000003.1300002.00000
50%0.000000580.0000003.3950002.00000
75%1.000000660.0000003.6700003.00000
max1.000000800.0000004.0000004.00000
\n", + "
" + ], + "text/plain": [ + " admit gre gpa rank\n", + "count 400.000000 400.000000 400.000000 400.00000\n", + "mean 0.317500 587.700000 3.389900 2.48500\n", + "std 0.466087 115.516536 0.380567 0.94446\n", + "min 0.000000 220.000000 2.260000 1.00000\n", + "25% 0.000000 520.000000 3.130000 2.00000\n", + "50% 0.000000 580.000000 3.395000 2.00000\n", + "75% 1.000000 660.000000 3.670000 3.00000\n", + "max 1.000000 800.000000 4.000000 4.00000" + ] + }, + "execution_count": 69, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df.describe()" + ] + }, + { + "cell_type": "code", + "execution_count": 70, + "metadata": {}, + "outputs": [], + "source": [ + "df_ct = pd.crosstab(index=[df['admit']], columns=[df['rank']], margins=True) " + ] + }, + { + "cell_type": "code", + "execution_count": 71, + "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", + "
rank1234All
admit
028979355273
133542812127
All6115112167400
\n", + "
" + ], + "text/plain": [ + "rank 1 2 3 4 All\n", + "admit \n", + "0 28 97 93 55 273\n", + "1 33 54 28 12 127\n", + "All 61 151 121 67 400" + ] + }, + "execution_count": 71, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df_ct" + ] + }, + { + "cell_type": "code", + "execution_count": 72, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 72, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAEFCAYAAAACFke6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3KwsAAAAAElFTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "df.plot(kind=\"bar\")\n" + ] + }, + { + "cell_type": "code", + "execution_count": 73, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 73, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAD8CAYAAAB6paOMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "df.plot( x='gpa', y='gre', kind='hist')" + ] + }, + { + "cell_type": "code", + "execution_count": 80, + "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", + "
CoefColumnsIntScore
0[[0.00139291052464]]gre[-1.54917610063]0.681818
1[[0.319303457297]]gpa[-1.81404627438]0.681818
2[[-0.504582785938]]rank[0.429516683081]0.681818
3[[0.00124646717199, 0.12683616908]]gre, gpa[-1.891342199]0.681818
4[[0.00178403124724, -0.53347199398]]gre, rank[-0.536909864111]0.734848
\n", + "
" + ], + "text/plain": [ + " Coef Columns Int \\\n", + "0 [[0.00139291052464]] gre [-1.54917610063] \n", + "1 [[0.319303457297]] gpa [-1.81404627438] \n", + "2 [[-0.504582785938]] rank [0.429516683081] \n", + "3 [[0.00124646717199, 0.12683616908]] gre, gpa [-1.891342199] \n", + "4 [[0.00178403124724, -0.53347199398]] gre, rank [-0.536909864111] \n", + "\n", + " Score \n", + "0 0.681818 \n", + "1 0.681818 \n", + "2 0.681818 \n", + "3 0.681818 \n", + "4 0.734848 " + ] + }, + "execution_count": 80, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "rows = []\n", + "for i in range(1,11):\n", + " combos = list(combinations(['gre', 'gpa', 'rank'],i))\n", + " for j,com in enumerate(combos):\n", + " y = df.admit\n", + " X = pd.DataFrame(df, columns=com)\n", + " X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)\n", + " model = LogisticRegression().fit(X_train, y_train)\n", + " score = model.score(X_test, y_test)\n", + " s = ', '.join(com)\n", + " rows.append({'Score':score, 'Columns':s, 'Coef':model.coef_,'Int':model.intercept_})\n", + " df1 = pd.DataFrame(rows)\n", + "df1.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 81, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "y_hat = model.predict(X_test) " + ] + }, + { + "cell_type": "code", + "execution_count": 82, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[90, 0],\n", + " [32, 10]], dtype=int64)" + ] + }, + "execution_count": 82, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "confusion_matrix(y_test, y_hat)" + ] + }, + { + "cell_type": "code", + "execution_count": 83, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "0.75757575757575757" + ] + }, + "execution_count": 83, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "model.score(X_test, y_test)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "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": "3.6.2" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/grad_school.ipynb b/grad_school.ipynb new file mode 100644 index 0000000..93ce608 --- /dev/null +++ b/grad_school.ipynb @@ -0,0 +1,585 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 66, + "metadata": {}, + "outputs": [], + "source": [ + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "from sklearn.linear_model import LogisticRegression\n", + "from sklearn.metrics import confusion_matrix\n", + "from sklearn.model_selection import train_test_split\n", + "from pandas.plotting import scatter_matrix\n", + "from itertools import combinations\n", + "from sklearn import linear_model\n", + "from sklearn.model_selection import cross_val_score\n", + "%matplotlib inline" + ] + }, + { + "cell_type": "code", + "execution_count": 67, + "metadata": {}, + "outputs": [], + "source": [ + "df = pd.read_csv('grad.csv')" + ] + }, + { + "cell_type": "code", + "execution_count": 68, + "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", + "
admitgregparank
003803.613
116603.673
218004.001
316403.194
405202.934
\n", + "
" + ], + "text/plain": [ + " admit gre gpa rank\n", + "0 0 380 3.61 3\n", + "1 1 660 3.67 3\n", + "2 1 800 4.00 1\n", + "3 1 640 3.19 4\n", + "4 0 520 2.93 4" + ] + }, + "execution_count": 68, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "y = df.admit\n", + "X = df[['gpa','gre','rank']]\n", + "df.head(5)" + ] + }, + { + "cell_type": "code", + "execution_count": 69, + "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", + "
admitgregparank
count400.000000400.000000400.000000400.00000
mean0.317500587.7000003.3899002.48500
std0.466087115.5165360.3805670.94446
min0.000000220.0000002.2600001.00000
25%0.000000520.0000003.1300002.00000
50%0.000000580.0000003.3950002.00000
75%1.000000660.0000003.6700003.00000
max1.000000800.0000004.0000004.00000
\n", + "
" + ], + "text/plain": [ + " admit gre gpa rank\n", + "count 400.000000 400.000000 400.000000 400.00000\n", + "mean 0.317500 587.700000 3.389900 2.48500\n", + "std 0.466087 115.516536 0.380567 0.94446\n", + "min 0.000000 220.000000 2.260000 1.00000\n", + "25% 0.000000 520.000000 3.130000 2.00000\n", + "50% 0.000000 580.000000 3.395000 2.00000\n", + "75% 1.000000 660.000000 3.670000 3.00000\n", + "max 1.000000 800.000000 4.000000 4.00000" + ] + }, + "execution_count": 69, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df.describe()" + ] + }, + { + "cell_type": "code", + "execution_count": 70, + "metadata": {}, + "outputs": [], + "source": [ + "df_ct = pd.crosstab(index=[df['admit']], columns=[df['rank']], margins=True) " + ] + }, + { + "cell_type": "code", + "execution_count": 71, + "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", + "
rank1234All
admit
028979355273
133542812127
All6115112167400
\n", + "
" + ], + "text/plain": [ + "rank 1 2 3 4 All\n", + "admit \n", + "0 28 97 93 55 273\n", + "1 33 54 28 12 127\n", + "All 61 151 121 67 400" + ] + }, + "execution_count": 71, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "df_ct" + ] + }, + { + "cell_type": "code", + "execution_count": 72, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 72, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXsAAAEFCAYAAAACFke6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3KwsAAAAAElFTkSuQmCC\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "df.plot(kind=\"bar\")\n" + ] + }, + { + "cell_type": "code", + "execution_count": 73, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 73, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAD8CAYAAAB6paOMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAFhVJREFUeJzt3X2wXHWd5/H3RxINIPKUwEbCGNiiGK0pwXB1UVhqB3xG\nE6ZKR9TSrIVmHdkdH2smulMjlk4VbrnKWG6J2WFmMw4ICCgIzihEdEvLAi8QBRNdFJl4BcmVGQyI\nEaLf/aPPxTvxJrfzcLpv3/N+VXX1Ob8+p8/3l3T3557nVBWSpO560rALkCQNl0EgSR1nEEhSxxkE\nktRxBoEkdZxBIEkdZxBIUscZBJLUcQaBJHXcgmEX0I/FixfX8uXLh12GJI2U22677WdVtWS26UYi\nCJYvX874+Piwy5CkkZLkn/uZzk1DktRxBoEkdZxBIEkdNxL7CCRpf3r88ceZmJhg+/btwy5lv1i0\naBHLli1j4cKFezW/QSCpcyYmJjjkkENYvnw5SYZdzj6pKh588EEmJiY47rjj9uo93DQkqXO2b9/O\nkUceOfIhAJCEI488cp/WbgwCSZ00H0Jgyr72xSCQpI5zH4Gkzlu+9ob9+n73Xnj2fn2/thkE0oja\n3z9e/Rq1H7lRtWPHDhYsGMxPdKtLSfJO4M1AAXcCbwKWApcDRwC3A2+oqsfarEOS5poPfvCDXHrp\npRx77LEsXryYU045heuvv54XvOAFfOMb32DlypW88Y1v5K1vfStbtmwB4KKLLuK0007b77W0FgRJ\njgH+FHhWVf0yyZXAucDLgY9V1eVJLgbOAz7ZVh2SNNeMj49z9dVXc8cdd7Bjxw5WrFjBKaecAsBD\nDz3E1772NQBe97rX8c53vpPTTz+dLVu28JKXvITNmzfv93raXu9YAByY5HHgIOB+4Ezgdc3r64EL\nMAgkdcjXv/51Vq1axYEHHgjAK1/5yidee81rXvPE8E033cSmTZueGN+2bRsPP/wwhxxyyH6tp7Ug\nqKqfJPkIsAX4JfBl4Dbgoara0Uw2ARzTVg2SNBdV1S5fO/jgg58Y/s1vfsM3v/nNJwKjLa0dPprk\ncGAVcBzwdOBg4GUzTDrjv0iSNUnGk4xPTk62VaYkDdzpp5/OF77wBbZv384jjzzCDTfMvOP/xS9+\nMZ/4xCeeGN+4cWMr9bS5aeiFwI+qahIgyTXAC4DDkixo1gqWAffNNHNVrQPWAYyNje06PiVpHw36\nSKjnPve5rFy5kpNOOolnPOMZjI2Nceihh/7OdB//+Mc5//zzefazn82OHTs444wzuPjii/d7PW0G\nwRbg1CQH0ds0dBYwDtwMvIrekUOrgWtbrEGS5qT3vOc9XHDBBTz66KOcccYZvPvd7+Ytb3nLv5lm\n8eLFXHHFFa3X0uY+gluSXEXvENEdwB30/sK/Abg8yYeatkvaqkGS5qo1a9awadMmtm/fzurVq1mx\nYsXQamn1qKGqej/w/p2a7wGe1+ZyJWmuu+yyy4ZdwhO81pCkTtrdkTujZl/7YhBI6pxFixbx4IMP\nzoswmLofwaJFi/b6PbzWkKTOWbZsGRMTE8yXQ9On7lC2twwCSZ2zcOHCvb6b13zkpiFJ6jiDQJI6\nziCQpI5zH4G0D4Z1cxhpf3KNQJI6ziCQpI4zCCSp4wwCSeo4g0CSOs4gkKSOMwgkqeMMAknquDZv\nXn9iko3THtuSvCPJEUluTHJ383x4WzVIkmbXWhBU1fer6uSqOhk4BXgU+BywFthQVScAG5pxSdKQ\nDGrT0FnAD6vqn4FVwPqmfT1wzoBqkCTNYFBBcC7wmWb46Kq6H6B5PmpANUiSZtB6ECR5MrAS+Owe\nzrcmyXiS8flyFyFJmosGsUbwMuD2qnqgGX8gyVKA5nnrTDNV1bqqGquqsSVLlgygTEnqpkEEwWv5\n7WYhgOuA1c3wauDaAdQgSdqFVoMgyUHAi4BrpjVfCLwoyd3Naxe2WYMkafdavTFNVT0KHLlT24P0\njiKSJM0BnlksSR1nEEhSxxkEktRxBoEkdZxBIEkdZxBIUscZBJLUcQaBJHVcqyeUSZp/lq+9YWjL\nvvfCs4e27PnMNQJJ6jiDQJI6ziCQpI4zCCSp4wwCSeo4g0CSOs4gkKSOMwgkqePavlXlYUmuSvK9\nJJuTPD/JEUluTHJ383x4mzVIknav7TWCvwb+qap+HzgJ2AysBTZU1QnAhmZckjQkrQVBkqcBZwCX\nAFTVY1X1ELAKWN9Mth44p60aJEmza3ON4HhgEvi7JHck+ZskBwNHV9X9AM3zUTPNnGRNkvEk45OT\nky2WKUnd1mYQLABWAJ+squcAv2APNgNV1bqqGquqsSVLlrRVoyR1XptBMAFMVNUtzfhV9ILhgSRL\nAZrnrS3WIEmaRWtBUFU/BX6c5MSm6SxgE3AdsLppWw1c21YNkqTZtX0/gv8GXJrkycA9wJvohc+V\nSc4DtgCvbrkGSdJutBoEVbURGJvhpbPaXK4kqX+eWSxJHWcQSFLHGQSS1HEGgSR1nEEgSR1nEEhS\nxxkEktRxBoEkdZxBIEkdZxBIUscZBJLUcQaBJHWcQSBJHddXECT5g7YLkSQNR79rBBcnuTXJ25Ic\n1mpFkqSB6isIqup04PXAscB4ksuSvKjVyiRJA9H3jWmq6u4kfwGMAx8HnpMkwPuq6pqZ5klyL/Aw\n8GtgR1WNJTkCuAJYDtwL/HFV/eu+dEKStPf6CoIkz6Z3m8mzgRuBV1bV7UmeDnwTmDEIGn9YVT+b\nNr4W2FBVFyZZ24z/+V5VLzWWr71h2CVII6vffQSfAG4HTqqq86vqdoCqug/4iz1c5ipgfTO8Hjhn\nD+eXJO1H/W4aejnwy6r6NUCSJwGLqurRqvr0buYr4MtJCvhUVa0Djq6q+wGq6v4kR+1D/ZKkfdTv\nGsFNwIHTxg9q2mZzWlWtAF4GnJ/kjH4LS7ImyXiS8cnJyX5nkyTtoX6DYFFVPTI10gwfNNtMzaYj\nqmor8DngecADSZYCNM9bdzHvuqoaq6qxJUuW9FmmJGlP9RsEv0iyYmokySnAL3c3Q5KDkxwyNQy8\nGLgLuA5Y3Uy2Grh2T4uWJO0//e4jeAfw2ST3NeNLgdfMMs/RwOd6R5iyALisqv4pybeAK5OcB2wB\nXr3nZUuS9pe+gqCqvpXk94ETgQDfq6rHZ5nnHuCkGdofBM7ai1olSS3o+4Qy4Ln0TgJbQO9kMqrq\n71upSpI0MP2eUPZp4N8DG+mdJQy9Q0MNAkkacf2uEYwBz6qqarMYSdLg9XvU0F3Av2uzEEnScPS7\nRrAY2JTkVuBXU41VtbKVqiRJA9NvEFzQZhGSpOHp9/DRryV5BnBCVd2U5CDggHZLkyQNQr+3qnwL\ncBXwqabpGODzbRUlSRqcfncWnw+cBmyD3k1qAK8aKknzQL9B8KuqemxqJMkCeucRSJJGXL87i7+W\n5H3Agc29it8GfKG9siTpdw3rTnT3Xnj2UJY7KP2uEawFJoE7gf8CfJE9vzOZJGkO6veood8A/7t5\nSJLmkX6vNfQjZtgnUFXH7/eKJEkDtSfXGpqyiN49BI7Y/+VIkgatr30EVfXgtMdPquoi4MyWa5Mk\nDUC/m4ZWTBt9Er01hEP6nPcAYBz4SVW9IslxwOX01ihuB94w/dBUSdJg9btp6H9OG94B3Av8cZ/z\nvh3YDDytGf8w8LGqujzJxcB5wCf7fC9J0n7W71FDf7g3b55kGXA28FfAu9K7gfGZwOuaSdbTu6Cd\nQSBJQ9LvpqF37e71qvroLl66CPgzfrsZ6Ujgoara0YxP0LtukSRpSPo9oWwM+BN6P9rHAG8FnkXv\nB37GfQVJXgFsrarbpjfPMOmMl6pIsibJeJLxycnJPsuUJO2pPbkxzYqqehggyQXAZ6vqzbuZ5zRg\nZZKX0zvk9Gn01hAOS7KgWStYBtw308xVtQ5YBzA2NuZ1jSSpJf2uEfweMP3InseA5buboareW1XL\nqmo5cC7wlap6PXAz8KpmstXAtXtSsCRp/+p3jeDTwK1JPkdvU84fAX+/l8v8c+DyJB8C7gAu2cv3\nkSTtB/0eNfRXSf4R+I9N05uq6o5+F1JVXwW+2gzfAzxvz8qUJLWl301DAAcB26rqr4GJ5sQwSdKI\n6/dWle+nt0nnvU3TQuAf2ipKkjQ4/a4R/BGwEvgFQFXdR5+XmJAkzW39BsFjVVU0x/wnObi9kiRJ\ng9RvEFyZ5FP0zgF4C3AT3qRGkuaFfo8a+khzr+JtwInAX1bVja1WJkkaiFmDoLmM9Jeq6oWAP/6S\nNM/MGgRV9eskjyY5tKp+PoiiJGkuWb72hqEs994Lzx7Icvo9s3g7cGeSG2mOHAKoqj9tpSpJ0sD0\nGwQ3NA9J0jyz2yBI8ntVtaWq1g+qIEnSYM12+OjnpwaSXN1yLZKkIZgtCKbfSOb4NguRJA3HbEFQ\nuxiWJM0Ts+0sPinJNnprBgc2wzTjVVVPa7U6SVLrdhsEVXXAoAqRJA3HntyPYI8kWZTk1iTfTvLd\nJB9o2o9LckuSu5NckeTJbdUgSZpda0EA/Ao4s6pOAk4GXprkVODDwMeq6gTgX4HzWqxBkjSL1oKg\neh5pRhc2jwLOBK5q2tcD57RVgyRpdm2uEZDkgCQbga30Llj3Q+ChqtrRTDIBHNNmDZKk3Ws1CKrq\n11V1MrCM3g3rnznTZDPNm2RNkvEk45OTk22WKUmd1moQTKmqh4CvAqfSu7nN1NFKy4D7djHPuqoa\nq6qxJUuWDKJMSeqkNo8aWpLksGb4QOCFwGbgZuBVzWSrgWvbqkGSNLt+rz66N5YC65sb2zwJuLKq\nrk+yCbg8yYeAO4BLWqxBkjSL1oKgqr4DPGeG9nvo7S+QJM0Bba4RqGOGdRcnSftmIDuLJUlzl0Eg\nSR1nEEhSxxkEktRxBoEkdZxBIEkdZxBIUscZBJLUcQaBJHWcQSBJHWcQSFLHGQSS1HEGgSR1nEEg\nSR1nEEhSxxkEktRxbd6z+NgkNyfZnOS7Sd7etB+R5MYkdzfPh7dVgyRpdm2uEewA3l1VzwROBc5P\n8ixgLbChqk4ANjTjkqQhaS0Iqur+qrq9GX4Y2AwcA6wC1jeTrQfOaasGSdLsBrKPIMlyejeyvwU4\nuqruh15YAEftYp41ScaTjE9OTg6iTEnqpNaDIMlTgauBd1TVtn7nq6p1VTVWVWNLlixpr0BJ6rhW\ngyDJQnohcGlVXdM0P5BkafP6UmBrmzVIknavzaOGAlwCbK6qj0576TpgdTO8Gri2rRokSbNb0OJ7\nnwa8Abgzycam7X3AhcCVSc4DtgCvbrEGSdIsWguCqvo6kF28fFZby5Uk7RnPLJakjjMIJKnjDAJJ\n6jiDQJI6rs2jhjQky9feMOwSJI0Q1wgkqeMMAknqOINAkjrOIJCkjjMIJKnjDAJJ6jiDQJI6ziCQ\npI4zCCSp4wwCSeo4g0CSOq7NW1X+bZKtSe6a1nZEkhuT3N08H97W8iVJ/WlzjeD/AC/dqW0tsKGq\nTgA2NOOSpCFqLQiq6v8C/7JT8ypgfTO8HjinreVLkvoz6H0ER1fV/QDN81EDXr4kaSdzdmdxkjVJ\nxpOMT05ODrscSZq3Bh0EDyRZCtA8b93VhFW1rqrGqmpsyZIlAytQkrpm0EFwHbC6GV4NXDvg5UuS\ndtLm4aOfAb4JnJhkIsl5wIXAi5LcDbyoGZckDVFr9yyuqtfu4qWz2lqmJGnPzdmdxZKkwTAIJKnj\nDAJJ6jiDQJI6ziCQpI4zCCSp4wwCSeo4g0CSOs4gkKSOMwgkqeMMAknqOINAkjqutYvOdd3ytTcM\nuwRJ6otrBJLUcQaBJHWcQSBJHTeUIEjy0iTfT/KDJGuHUYMkqWfgO4uTHAD8L3q3qpwAvpXkuqra\n1Mby3GkrSbs3jDWC5wE/qKp7quox4HJg1RDqkCQxnCA4BvjxtPGJpk2SNATDOI8gM7TV70yUrAHW\nNKOPJPn+TpMsBn62n2sbpvnWH5h/fbI/c9+86lM+vM/9eUY/Ew0jCCaAY6eNLwPu23miqloHrNvV\nmyQZr6qx/V/ecMy3/sD865P9mfvmW58G1Z9hbBr6FnBCkuOSPBk4F7huCHVIkhjCGkFV7UjyX4Ev\nAQcAf1tV3x10HZKknqFca6iqvgh8cR/fZpebjUbUfOsPzL8+2Z+5b771aSD9SdXv7KeVJHWIl5iQ\npI6bk0GQ5NgkNyfZnOS7Sd7etB+R5MYkdzfPhzftSfLx5pIV30myYrg9+LeSLEpya5JvN/35QNN+\nXJJbmv5c0ew8J8lTmvEfNK8vH2b9u5PkgCR3JLm+GR/ZPiW5N8mdSTYmGW/aRvIzNyXJYUmuSvK9\n5vv0/FHtU5ITm/+bqce2JO8Y1f5MSfLO5nfhriSfaX4vBvs9qqo59wCWAiua4UOA/wc8C/gfwNqm\nfS3w4Wb45cA/0jtH4VTglmH3Yaf+BHhqM7wQuKWp80rg3Kb9YuBPmuG3ARc3w+cCVwy7D7vp27uA\ny4Drm/GR7RNwL7B4p7aR/MxNq3898OZm+MnAYaPep6bWA4Cf0jtOfmT7Q+9k2h8BBzbjVwL/edDf\no6H/Q/T5j3UtvWsTfR9Y2rQtBb7fDH8KeO206Z+Ybq49gIOA24H/QO9EkQVN+/OBLzXDXwKe3wwv\naKbLsGufoS/LgA3AmcD1zRduZPu0iyAY2c8c8LTmRyY7tY9sn6bV9mLgG6PeH357pYUjmu/F9cBL\nBv09mpObhqZrVn2eQ++v6KOr6n6A5vmoZrI5f9mKZhPKRmArcCPwQ+ChqtrRTDK95if607z+c+DI\nwVbcl4uAPwN+04wfyWj3qYAvJ7ktvTPbYYQ/c8DxwCTwd83mu79JcjCj3acp5wKfaYZHtj9V9RPg\nI8AW4H5634vbGPD3aE4HQZKnAlcD76iqbbubdIa2OXU4VFX9uqpOpvdX9POAZ840WfM85/uT5BXA\n1qq6bXrzDJOOTJ+A06pqBfAy4PwkZ+xm2lHozwJgBfDJqnoO8At6m052ZRT6RLO9fCXw2dkmnaFt\nTvWn2Z+xCjgOeDpwML3P385a/R7N2SBIspBeCFxaVdc0zQ8kWdq8vpTeX9fQ52Ur5oKqegj4Kr1t\nloclmTqXY3rNT/Snef1Q4F8GW+msTgNWJrmX3hVkz6S3hjCyfaqq+5rnrcDn6AX2KH/mJoCJqrql\nGb+KXjCMcp+g90N5e1U90IyPcn9eCPyoqiar6nHgGuAFDPh7NCeDIEmAS4DNVfXRaS9dB6xuhlfT\n23cw1f7G5iiBU4GfT60qzgVJliQ5rBk+kN5//mbgZuBVzWQ792eqn68CvlLNRsG5oqreW1XLqmo5\nvdX0r1TV6xnRPiU5OMkhU8P0tkHfxYh+5gCq6qfAj5Oc2DSdBWxihPvUeC2/3SwEo92fLcCpSQ5q\nfvem/o8G+z0a9s6SXexAOZ3e6s53gI3N4+X0toVtAO5uno9opg+9m938ELgTGBt2H3bqz7OBO5r+\n3AX8ZdN+PHAr8AN6q7lPadoXNeM/aF4/fth9mKV//4nfHjU0kn1q6v528/gu8N+b9pH8zE3r18nA\nePPZ+zxw+Cj3id7BFg8Ch05rG9n+NHV+APhe89vwaeApg/4eeWaxJHXcnNw0JEkaHINAkjrOIJCk\njjMIJKnjDAJJ6jiDQJI6ziCQpI4zCCSp4/4/h4PxzFB9+GQAAAAASUVORK5CYII=\n", + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "df.plot( x='gpa', y='gre', kind='hist')" + ] + }, + { + "cell_type": "code", + "execution_count": 80, + "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", + "
CoefColumnsIntScore
0[[0.00139291052464]]gre[-1.54917610063]0.681818
1[[0.319303457297]]gpa[-1.81404627438]0.681818
2[[-0.504582785938]]rank[0.429516683081]0.681818
3[[0.00124646717199, 0.12683616908]]gre, gpa[-1.891342199]0.681818
4[[0.00178403124724, -0.53347199398]]gre, rank[-0.536909864111]0.734848
\n", + "
" + ], + "text/plain": [ + " Coef Columns Int \\\n", + "0 [[0.00139291052464]] gre [-1.54917610063] \n", + "1 [[0.319303457297]] gpa [-1.81404627438] \n", + "2 [[-0.504582785938]] rank [0.429516683081] \n", + "3 [[0.00124646717199, 0.12683616908]] gre, gpa [-1.891342199] \n", + "4 [[0.00178403124724, -0.53347199398]] gre, rank [-0.536909864111] \n", + "\n", + " Score \n", + "0 0.681818 \n", + "1 0.681818 \n", + "2 0.681818 \n", + "3 0.681818 \n", + "4 0.734848 " + ] + }, + "execution_count": 80, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "rows = []\n", + "for i in range(1,11):\n", + " combos = list(combinations(['gre', 'gpa', 'rank'],i))\n", + " for j,com in enumerate(combos):\n", + " y = df.admit\n", + " X = pd.DataFrame(df, columns=com)\n", + " X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)\n", + " model = LogisticRegression().fit(X_train, y_train)\n", + " score = model.score(X_test, y_test)\n", + " s = ', '.join(com)\n", + " rows.append({'Score':score, 'Columns':s, 'Coef':model.coef_,'Int':model.intercept_})\n", + " df1 = pd.DataFrame(rows)\n", + "df1.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 81, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "y_hat = model.predict(X_test) " + ] + }, + { + "cell_type": "code", + "execution_count": 82, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "array([[90, 0],\n", + " [32, 10]], dtype=int64)" + ] + }, + "execution_count": 82, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "confusion_matrix(y_test, y_hat)" + ] + }, + { + "cell_type": "code", + "execution_count": 83, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "0.75757575757575757" + ] + }, + "execution_count": 83, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "model.score(X_test, y_test)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "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": "3.6.2" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +}