{"id":423,"date":"2024-02-29T14:33:48","date_gmt":"2024-02-29T19:33:48","guid":{"rendered":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/chapter\/decision-trees-heloc_2024\/"},"modified":"2024-02-29T14:35:36","modified_gmt":"2024-02-29T19:35:36","slug":"decision-trees-heloc_2024","status":"publish","type":"chapter","link":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/chapter\/decision-trees-heloc_2024\/","title":{"raw":"Decision Trees in Python","rendered":"Decision Trees in Python"},"content":{"raw":"
\r\n
\r\n
<\/div>\r\n
\r\n
\r\n

Decision Trees\u00b6<\/a><\/h2>\r\nIn which we talk about decision trees!\r\n\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[1]:<\/div>\r\n
\r\n
\r\n
\r\n
import<\/span> numpy<\/span> as<\/span> np<\/span>\r\nimport<\/span> pandas<\/span>\r\nimport<\/span> matplotlib.pyplot<\/span> as<\/span> plt<\/span>\r\n%<\/span>matplotlib<\/span> inline\r\nimport<\/span> os<\/span>\r\n\r\nfrom<\/span> sklearn.model_selection<\/span> import<\/span> train_test_split<\/span>\r\nfrom<\/span> sklearn<\/span> import<\/span> datasets<\/span>\r\n\r\n##New!<\/span>\r\nfrom<\/span> sklearn<\/span> import<\/span> tree<\/span>\r\n\r\n\r\n\r\n#os.getcwd() ## run if things aren't working as expected!<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[2]:<\/div>\r\n
\r\n
\r\n
\r\n
import<\/span> sklearn<\/span> as<\/span> sklearn<\/span>\r\n\r\nsklearn<\/span>.<\/span>__version__<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
<\/div>\r\n
\r\n
\r\n
Out[2]:<\/div>\r\n
\r\n
'1.2.1'<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[3]:<\/div>\r\n
\r\n
\r\n
\r\n
##load our dataset<\/span>\r\nheloc<\/span>=<\/span> pandas<\/span>.<\/span>read_csv<\/span>(<\/span>\"data\\Heloc_labeled.csv\"<\/span>)<\/span> \r\n\r\n## look it over<\/span>\r\nheloc<\/span>.<\/span>head<\/span>(<\/span>10<\/span>)<\/span>\r\n#heloc.shape<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
<\/div>\r\n
\r\n
\r\n
Out[3]:<\/div>\r\n
\r\n
\r\n\r\n.dataframe tbody tr th:only-of-type {\r\nvertical-align: middle;\r\n}\r\n\r\n.dataframe tbody tr th {\r\nvertical-align: top;\r\n}\r\n\r\n.dataframe thead th {\r\ntext-align: right;\r\n}\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n
<\/th>\r\nAge<\/th>\r\nSex<\/th>\r\nIncome<\/th>\r\nHELOC<\/th>\r\n<\/tr>\r\n<\/thead>\r\n
0<\/th>\r\n30<\/td>\r\nFemale<\/td>\r\n101000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
1<\/th>\r\n25<\/td>\r\nMale<\/td>\r\n86000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
2<\/th>\r\n20<\/td>\r\nMale<\/td>\r\n50000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
3<\/th>\r\n26<\/td>\r\nMale<\/td>\r\n58000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
4<\/th>\r\n18<\/td>\r\nFemale<\/td>\r\n93000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
5<\/th>\r\n38<\/td>\r\nMale<\/td>\r\n153000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
6<\/th>\r\n61<\/td>\r\nMale<\/td>\r\n71000<\/td>\r\n1<\/td>\r\n<\/tr>\r\n
7<\/th>\r\n27<\/td>\r\nMale<\/td>\r\n102000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
8<\/th>\r\n38<\/td>\r\nMale<\/td>\r\n33000<\/td>\r\n1<\/td>\r\n<\/tr>\r\n
9<\/th>\r\n42<\/td>\r\nFemale<\/td>\r\n69000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[4]:<\/div>\r\n
\r\n
\r\n
\r\n
heloc<\/span>.<\/span>shape<\/span>[<\/span>1<\/span>]<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
<\/div>\r\n
\r\n
\r\n
Out[4]:<\/div>\r\n
\r\n
4<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[5]:<\/div>\r\n
\r\n
\r\n
\r\n
## That Male\/ Female split is going to be a problem<\/span>\r\n## Our data needs preprocessing:<\/span>\r\n\r\n## There are lots of ways to do this!<\/span>\r\n\r\ndef<\/span> sex_recode<\/span>(<\/span>sex<\/span>):<\/span>\r\n    \"\"\"<\/span>\r\n    everything inside the quotes is documentation<\/span>\r\n    \"\"\"<\/span>\r\n    if<\/span> sex<\/span> ==<\/span>\"Female\"<\/span>:<\/span>\r\n        return<\/span> 1<\/span>\r\n    elif<\/span> sex<\/span> ==<\/span> \"Male\"<\/span>:<\/span>\r\n        return<\/span> 0<\/span>\r\n    else<\/span>:<\/span>\r\n        return<\/span> 2<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[6]:<\/div>\r\n
\r\n
\r\n
\r\n
sex_recode<\/span>(<\/span>\"amy\"<\/span>)<\/span> ## TEst<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
<\/div>\r\n
\r\n
\r\n
Out[6]:<\/div>\r\n
\r\n
2<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[7]:<\/div>\r\n
\r\n
\r\n
\r\n
heloc<\/span>[<\/span>'Sex'<\/span>]<\/span>=<\/span>heloc<\/span>[<\/span>'Sex'<\/span>]<\/span>.<\/span>apply<\/span>(<\/span>sex_recode<\/span>)<\/span>\r\n\r\n\r\n\r\nheloc<\/span>.<\/span>head<\/span>(<\/span>5<\/span>)<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
<\/div>\r\n
\r\n
\r\n
Out[7]:<\/div>\r\n
\r\n
\r\n\r\n.dataframe tbody tr th:only-of-type {\r\nvertical-align: middle;\r\n}\r\n\r\n.dataframe tbody tr th {\r\nvertical-align: top;\r\n}\r\n\r\n.dataframe thead th {\r\ntext-align: right;\r\n}\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n
<\/th>\r\nAge<\/th>\r\nSex<\/th>\r\nIncome<\/th>\r\nHELOC<\/th>\r\n<\/tr>\r\n<\/thead>\r\n
0<\/th>\r\n30<\/td>\r\n1<\/td>\r\n101000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
1<\/th>\r\n25<\/td>\r\n0<\/td>\r\n86000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
2<\/th>\r\n20<\/td>\r\n0<\/td>\r\n50000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
3<\/th>\r\n26<\/td>\r\n0<\/td>\r\n58000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
4<\/th>\r\n18<\/td>\r\n1<\/td>\r\n93000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[8]:<\/div>\r\n
\r\n
\r\n
\r\n
## Let's split the X and Y:<\/span>\r\n\r\nX<\/span> =<\/span> heloc<\/span>[[<\/span>'Age'<\/span>,<\/span> 'Sex'<\/span>,<\/span> 'Income'<\/span>]]<\/span>\r\nY<\/span> =<\/span> heloc<\/span>[[<\/span>'HELOC'<\/span>]]<\/span>\r\n\r\n\r\nY<\/span>.<\/span>tail<\/span>()<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
<\/div>\r\n
\r\n
\r\n
Out[8]:<\/div>\r\n
\r\n
\r\n\r\n.dataframe tbody tr th:only-of-type {\r\nvertical-align: middle;\r\n}\r\n\r\n.dataframe tbody tr th {\r\nvertical-align: top;\r\n}\r\n\r\n.dataframe thead th {\r\ntext-align: right;\r\n}\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n
<\/th>\r\nHELOC<\/th>\r\n<\/tr>\r\n<\/thead>\r\n
495<\/th>\r\n1<\/td>\r\n<\/tr>\r\n
496<\/th>\r\n1<\/td>\r\n<\/tr>\r\n
497<\/th>\r\n0<\/td>\r\n<\/tr>\r\n
498<\/th>\r\n0<\/td>\r\n<\/tr>\r\n
499<\/th>\r\n1<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[9]:<\/div>\r\n
\r\n
\r\n
\r\n
#and build a model<\/span>\r\n\r\ntree_heloc<\/span>=<\/span> tree<\/span>.<\/span>DecisionTreeClassifier<\/span>(<\/span>max_depth<\/span>=<\/span> 2<\/span>,<\/span> random_state<\/span> =<\/span> 2021<\/span>)<\/span>\r\ntree_heloc<\/span>=<\/span> tree_heloc<\/span>.<\/span>fit<\/span>(<\/span>X<\/span>,<\/span> Y<\/span>)<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[10]:<\/div>\r\n
\r\n
\r\n
\r\n
tree<\/span>.<\/span>plot_tree<\/span>(<\/span>tree_heloc<\/span>,<\/span> filled<\/span> =<\/span> True<\/span>,<\/span> feature_names<\/span> =<\/span> [<\/span>'Age'<\/span>,<\/span> 'Sex'<\/span>,<\/span> 'Income'<\/span>]);<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\"image\"<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[15]:<\/div>\r\n
\r\n
\r\n
\r\n
?<\/span>tree.DecisionTreeClassifier\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[12]:<\/div>\r\n
\r\n
\r\n
\r\n
## Let's predict on our data!<\/span>\r\nheloc_new<\/span>=<\/span> pandas<\/span>.<\/span>read_csv<\/span>(<\/span>\"data\\Heloc_unlabeled.csv\"<\/span>)<\/span> \r\n\r\n## fix that pesky Sex variable!<\/span>\r\nheloc_new<\/span>[<\/span>'Sex'<\/span>]<\/span>=<\/span>heloc_new<\/span>[<\/span>'Sex'<\/span>]<\/span>.<\/span>apply<\/span>(<\/span>sex_recode<\/span>)<\/span>\r\n\r\n## look it over<\/span>\r\nheloc_new<\/span>.<\/span>head<\/span>(<\/span>3<\/span>)<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
<\/div>\r\n
\r\n
\r\n
Out[12]:<\/div>\r\n
\r\n
\r\n\r\n.dataframe tbody tr th:only-of-type {\r\nvertical-align: middle;\r\n}\r\n\r\n.dataframe tbody tr th {\r\nvertical-align: top;\r\n}\r\n\r\n.dataframe thead th {\r\ntext-align: right;\r\n}\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n
<\/th>\r\nAge<\/th>\r\nSex<\/th>\r\nIncome<\/th>\r\n<\/tr>\r\n<\/thead>\r\n
0<\/th>\r\n25<\/td>\r\n1<\/td>\r\n45000<\/td>\r\n<\/tr>\r\n
1<\/th>\r\n23<\/td>\r\n0<\/td>\r\n22000<\/td>\r\n<\/tr>\r\n
2<\/th>\r\n50<\/td>\r\n1<\/td>\r\n17000<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[13]:<\/div>\r\n
\r\n
\r\n
\r\n
## Trying our model, and making a new set:<\/span>\r\nY_new<\/span> =<\/span> tree_heloc<\/span>.<\/span>predict<\/span>(<\/span>heloc_new<\/span>)<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
\r\n
<\/div>\r\n
\r\n
In\u00a0[14]:<\/div>\r\n
\r\n
\r\n
\r\n
## I like to make a new data.frame:<\/span>\r\nheloc_p<\/span> =<\/span> heloc_new<\/span>\r\n\r\nheloc_p<\/span>[<\/span>'Heloc'<\/span>]<\/span>=<\/span>Y_new<\/span> ## add in the new column of predictions<\/span>\r\n\r\n## view!<\/span>\r\nheloc_p<\/span>\r\n<\/pre>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n
\r\n
<\/div>\r\n
\r\n
\r\n
Out[14]:<\/div>\r\n
\r\n
\r\n\r\n.dataframe tbody tr th:only-of-type {\r\nvertical-align: middle;\r\n}\r\n\r\n.dataframe tbody tr th {\r\nvertical-align: top;\r\n}\r\n\r\n.dataframe thead th {\r\ntext-align: right;\r\n}\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n
<\/th>\r\nAge<\/th>\r\nSex<\/th>\r\nIncome<\/th>\r\nHeloc<\/th>\r\n<\/tr>\r\n<\/thead>\r\n
0<\/th>\r\n25<\/td>\r\n1<\/td>\r\n45000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
1<\/th>\r\n23<\/td>\r\n0<\/td>\r\n22000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
2<\/th>\r\n50<\/td>\r\n1<\/td>\r\n17000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
3<\/th>\r\n28<\/td>\r\n1<\/td>\r\n38000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
4<\/th>\r\n56<\/td>\r\n0<\/td>\r\n24000<\/td>\r\n1<\/td>\r\n<\/tr>\r\n
5<\/th>\r\n50<\/td>\r\n1<\/td>\r\n24000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
6<\/th>\r\n22<\/td>\r\n0<\/td>\r\n27000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
7<\/th>\r\n22<\/td>\r\n0<\/td>\r\n40000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
8<\/th>\r\n21<\/td>\r\n0<\/td>\r\n41000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
9<\/th>\r\n29<\/td>\r\n1<\/td>\r\n45000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
10<\/th>\r\n64<\/td>\r\n0<\/td>\r\n47000<\/td>\r\n1<\/td>\r\n<\/tr>\r\n
11<\/th>\r\n47<\/td>\r\n1<\/td>\r\n48000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
12<\/th>\r\n55<\/td>\r\n1<\/td>\r\n49000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
13<\/th>\r\n25<\/td>\r\n0<\/td>\r\n54000<\/td>\r\n1<\/td>\r\n<\/tr>\r\n
14<\/th>\r\n42<\/td>\r\n0<\/td>\r\n113000<\/td>\r\n1<\/td>\r\n<\/tr>\r\n
15<\/th>\r\n40<\/td>\r\n1<\/td>\r\n118000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
16<\/th>\r\n38<\/td>\r\n0<\/td>\r\n153000<\/td>\r\n1<\/td>\r\n<\/tr>\r\n
17<\/th>\r\n63<\/td>\r\n0<\/td>\r\n156000<\/td>\r\n1<\/td>\r\n<\/tr>\r\n
18<\/th>\r\n51<\/td>\r\n1<\/td>\r\n172000<\/td>\r\n0<\/td>\r\n<\/tr>\r\n
19<\/th>\r\n51<\/td>\r\n0<\/td>\r\n43000<\/td>\r\n1<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>\r\n<\/div>","rendered":"
\n
\n
<\/div>\n
\n
\n
Decision Trees\u00b6<\/a><\/h2>\n
In which we talk about decision trees!<\/p>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[1]:<\/div>\n
\n
\n
\n
import<\/span> numpy<\/span> as<\/span> np<\/span>\r\nimport<\/span> pandas<\/span>\r\nimport<\/span> matplotlib.pyplot<\/span> as<\/span> plt<\/span>\r\n%<\/span>matplotlib<\/span> inline\r\nimport<\/span> os<\/span>\r\n\r\nfrom<\/span> sklearn.model_selection<\/span> import<\/span> train_test_split<\/span>\r\nfrom<\/span> sklearn<\/span> import<\/span> datasets<\/span>\r\n\r\n##New!<\/span>\r\nfrom<\/span> sklearn<\/span> import<\/span> tree<\/span>\r\n\r\n\r\n\r\n#os.getcwd() ## run if things aren't working as expected!<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[2]:<\/div>\n
\n
\n
\n
import<\/span> sklearn<\/span> as<\/span> sklearn<\/span>\r\n\r\nsklearn<\/span>.<\/span>__version__<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
<\/div>\n
\n
\n
Out[2]:<\/div>\n
\n
'1.2.1'<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[3]:<\/div>\n
\n
\n
\n
##load our dataset<\/span>\r\nheloc<\/span>=<\/span> pandas<\/span>.<\/span>read_csv<\/span>(<\/span>\"data\\Heloc_labeled.csv\"<\/span>)<\/span> \r\n\r\n## look it over<\/span>\r\nheloc<\/span>.<\/span>head<\/span>(<\/span>10<\/span>)<\/span>\r\n#heloc.shape<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
<\/div>\n
\n
\n
Out[3]:<\/div>\n
\n
\n
.dataframe tbody tr th:only-of-type {
\nvertical-align: middle;
\n}<\/p>\n
.dataframe tbody tr th {
\nvertical-align: top;
\n}<\/p>\n
.dataframe thead th {
\ntext-align: right;
\n}<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
<\/th>\nAge<\/th>\nSex<\/th>\nIncome<\/th>\nHELOC<\/th>\n<\/tr>\n<\/thead>\n
0<\/th>\n30<\/td>\nFemale<\/td>\n101000<\/td>\n0<\/td>\n<\/tr>\n
1<\/th>\n25<\/td>\nMale<\/td>\n86000<\/td>\n0<\/td>\n<\/tr>\n
2<\/th>\n20<\/td>\nMale<\/td>\n50000<\/td>\n0<\/td>\n<\/tr>\n
3<\/th>\n26<\/td>\nMale<\/td>\n58000<\/td>\n0<\/td>\n<\/tr>\n
4<\/th>\n18<\/td>\nFemale<\/td>\n93000<\/td>\n0<\/td>\n<\/tr>\n
5<\/th>\n38<\/td>\nMale<\/td>\n153000<\/td>\n0<\/td>\n<\/tr>\n
6<\/th>\n61<\/td>\nMale<\/td>\n71000<\/td>\n1<\/td>\n<\/tr>\n
7<\/th>\n27<\/td>\nMale<\/td>\n102000<\/td>\n0<\/td>\n<\/tr>\n
8<\/th>\n38<\/td>\nMale<\/td>\n33000<\/td>\n1<\/td>\n<\/tr>\n
9<\/th>\n42<\/td>\nFemale<\/td>\n69000<\/td>\n0<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[4]:<\/div>\n
\n
\n
\n
heloc<\/span>.<\/span>shape<\/span>[<\/span>1<\/span>]<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
<\/div>\n
\n
\n
Out[4]:<\/div>\n
\n
4<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[5]:<\/div>\n
\n
\n
\n
## That Male\/ Female split is going to be a problem<\/span>\r\n## Our data needs preprocessing:<\/span>\r\n\r\n## There are lots of ways to do this!<\/span>\r\n\r\ndef<\/span> sex_recode<\/span>(<\/span>sex<\/span>):<\/span>\r\n    \"\"\"<\/span>\r\n    everything inside the quotes is documentation<\/span>\r\n    \"\"\"<\/span>\r\n    if<\/span> sex<\/span> ==<\/span>\"Female\"<\/span>:<\/span>\r\n        return<\/span> 1<\/span>\r\n    elif<\/span> sex<\/span> ==<\/span> \"Male\"<\/span>:<\/span>\r\n        return<\/span> 0<\/span>\r\n    else<\/span>:<\/span>\r\n        return<\/span> 2<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[6]:<\/div>\n
\n
\n
\n
sex_recode<\/span>(<\/span>\"amy\"<\/span>)<\/span> ## TEst<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
<\/div>\n
\n
\n
Out[6]:<\/div>\n
\n
2<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[7]:<\/div>\n
\n
\n
\n
heloc<\/span>[<\/span>'Sex'<\/span>]<\/span>=<\/span>heloc<\/span>[<\/span>'Sex'<\/span>]<\/span>.<\/span>apply<\/span>(<\/span>sex_recode<\/span>)<\/span>\r\n\r\n\r\n\r\nheloc<\/span>.<\/span>head<\/span>(<\/span>5<\/span>)<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
<\/div>\n
\n
\n
Out[7]:<\/div>\n
\n
\n
.dataframe tbody tr th:only-of-type {
\nvertical-align: middle;
\n}<\/p>\n
.dataframe tbody tr th {
\nvertical-align: top;
\n}<\/p>\n
.dataframe thead th {
\ntext-align: right;
\n}<\/p>\n\n\n\n\n\n\n\n\n\n
<\/th>\nAge<\/th>\nSex<\/th>\nIncome<\/th>\nHELOC<\/th>\n<\/tr>\n<\/thead>\n
0<\/th>\n30<\/td>\n1<\/td>\n101000<\/td>\n0<\/td>\n<\/tr>\n
1<\/th>\n25<\/td>\n0<\/td>\n86000<\/td>\n0<\/td>\n<\/tr>\n
2<\/th>\n20<\/td>\n0<\/td>\n50000<\/td>\n0<\/td>\n<\/tr>\n
3<\/th>\n26<\/td>\n0<\/td>\n58000<\/td>\n0<\/td>\n<\/tr>\n
4<\/th>\n18<\/td>\n1<\/td>\n93000<\/td>\n0<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[8]:<\/div>\n
\n
\n
\n
## Let's split the X and Y:<\/span>\r\n\r\nX<\/span> =<\/span> heloc<\/span>[[<\/span>'Age'<\/span>,<\/span> 'Sex'<\/span>,<\/span> 'Income'<\/span>]]<\/span>\r\nY<\/span> =<\/span> heloc<\/span>[[<\/span>'HELOC'<\/span>]]<\/span>\r\n\r\n\r\nY<\/span>.<\/span>tail<\/span>()<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
<\/div>\n
\n
\n
Out[8]:<\/div>\n
\n
\n
.dataframe tbody tr th:only-of-type {
\nvertical-align: middle;
\n}<\/p>\n
.dataframe tbody tr th {
\nvertical-align: top;
\n}<\/p>\n
.dataframe thead th {
\ntext-align: right;
\n}<\/p>\n\n\n\n\n\n\n\n\n\n
<\/th>\nHELOC<\/th>\n<\/tr>\n<\/thead>\n
495<\/th>\n1<\/td>\n<\/tr>\n
496<\/th>\n1<\/td>\n<\/tr>\n
497<\/th>\n0<\/td>\n<\/tr>\n
498<\/th>\n0<\/td>\n<\/tr>\n
499<\/th>\n1<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[9]:<\/div>\n
\n
\n
\n
#and build a model<\/span>\r\n\r\ntree_heloc<\/span>=<\/span> tree<\/span>.<\/span>DecisionTreeClassifier<\/span>(<\/span>max_depth<\/span>=<\/span> 2<\/span>,<\/span> random_state<\/span> =<\/span> 2021<\/span>)<\/span>\r\ntree_heloc<\/span>=<\/span> tree_heloc<\/span>.<\/span>fit<\/span>(<\/span>X<\/span>,<\/span> Y<\/span>)<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[10]:<\/div>\n
\n
\n
\n
tree<\/span>.<\/span>plot_tree<\/span>(<\/span>tree_heloc<\/span>,<\/span> filled<\/span> =<\/span> True<\/span>,<\/span> feature_names<\/span> =<\/span> [<\/span>'Age'<\/span>,<\/span> 'Sex'<\/span>,<\/span> 'Income'<\/span>]);<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
<\/div>\n
\n
\n
<\/div>\n
\"image\"<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[15]:<\/div>\n
\n
\n
\n
?<\/span>tree.DecisionTreeClassifier\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[12]:<\/div>\n
\n
\n
\n
## Let's predict on our data!<\/span>\r\nheloc_new<\/span>=<\/span> pandas<\/span>.<\/span>read_csv<\/span>(<\/span>\"data\\Heloc_unlabeled.csv\"<\/span>)<\/span> \r\n\r\n## fix that pesky Sex variable!<\/span>\r\nheloc_new<\/span>[<\/span>'Sex'<\/span>]<\/span>=<\/span>heloc_new<\/span>[<\/span>'Sex'<\/span>]<\/span>.<\/span>apply<\/span>(<\/span>sex_recode<\/span>)<\/span>\r\n\r\n## look it over<\/span>\r\nheloc_new<\/span>.<\/span>head<\/span>(<\/span>3<\/span>)<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
<\/div>\n
\n
\n
Out[12]:<\/div>\n
\n
\n
.dataframe tbody tr th:only-of-type {
\nvertical-align: middle;
\n}<\/p>\n
.dataframe tbody tr th {
\nvertical-align: top;
\n}<\/p>\n
.dataframe thead th {
\ntext-align: right;
\n}<\/p>\n\n\n\n\n\n\n\n
<\/th>\nAge<\/th>\nSex<\/th>\nIncome<\/th>\n<\/tr>\n<\/thead>\n
0<\/th>\n25<\/td>\n1<\/td>\n45000<\/td>\n<\/tr>\n
1<\/th>\n23<\/td>\n0<\/td>\n22000<\/td>\n<\/tr>\n
2<\/th>\n50<\/td>\n1<\/td>\n17000<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[13]:<\/div>\n
\n
\n
\n
## Trying our model, and making a new set:<\/span>\r\nY_new<\/span> =<\/span> tree_heloc<\/span>.<\/span>predict<\/span>(<\/span>heloc_new<\/span>)<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
<\/div>\n
\n
In\u00a0[14]:<\/div>\n
\n
\n
\n
## I like to make a new data.frame:<\/span>\r\nheloc_p<\/span> =<\/span> heloc_new<\/span>\r\n\r\nheloc_p<\/span>[<\/span>'Heloc'<\/span>]<\/span>=<\/span>Y_new<\/span> ## add in the new column of predictions<\/span>\r\n\r\n## view!<\/span>\r\nheloc_p<\/span>\r\n<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
<\/div>\n
\n
\n
Out[14]:<\/div>\n
\n
\n
.dataframe tbody tr th:only-of-type {
\nvertical-align: middle;
\n}<\/p>\n
.dataframe tbody tr th {
\nvertical-align: top;
\n}<\/p>\n
.dataframe thead th {
\ntext-align: right;
\n}<\/p>\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
<\/th>\nAge<\/th>\nSex<\/th>\nIncome<\/th>\nHeloc<\/th>\n<\/tr>\n<\/thead>\n
0<\/th>\n25<\/td>\n1<\/td>\n45000<\/td>\n0<\/td>\n<\/tr>\n
1<\/th>\n23<\/td>\n0<\/td>\n22000<\/td>\n0<\/td>\n<\/tr>\n
2<\/th>\n50<\/td>\n1<\/td>\n17000<\/td>\n0<\/td>\n<\/tr>\n
3<\/th>\n28<\/td>\n1<\/td>\n38000<\/td>\n0<\/td>\n<\/tr>\n
4<\/th>\n56<\/td>\n0<\/td>\n24000<\/td>\n1<\/td>\n<\/tr>\n
5<\/th>\n50<\/td>\n1<\/td>\n24000<\/td>\n0<\/td>\n<\/tr>\n
6<\/th>\n22<\/td>\n0<\/td>\n27000<\/td>\n0<\/td>\n<\/tr>\n
7<\/th>\n22<\/td>\n0<\/td>\n40000<\/td>\n0<\/td>\n<\/tr>\n
8<\/th>\n21<\/td>\n0<\/td>\n41000<\/td>\n0<\/td>\n<\/tr>\n
9<\/th>\n29<\/td>\n1<\/td>\n45000<\/td>\n0<\/td>\n<\/tr>\n
10<\/th>\n64<\/td>\n0<\/td>\n47000<\/td>\n1<\/td>\n<\/tr>\n
11<\/th>\n47<\/td>\n1<\/td>\n48000<\/td>\n0<\/td>\n<\/tr>\n
12<\/th>\n55<\/td>\n1<\/td>\n49000<\/td>\n0<\/td>\n<\/tr>\n
13<\/th>\n25<\/td>\n0<\/td>\n54000<\/td>\n1<\/td>\n<\/tr>\n
14<\/th>\n42<\/td>\n0<\/td>\n113000<\/td>\n1<\/td>\n<\/tr>\n
15<\/th>\n40<\/td>\n1<\/td>\n118000<\/td>\n0<\/td>\n<\/tr>\n
16<\/th>\n38<\/td>\n0<\/td>\n153000<\/td>\n1<\/td>\n<\/tr>\n
17<\/th>\n63<\/td>\n0<\/td>\n156000<\/td>\n1<\/td>\n<\/tr>\n
18<\/th>\n51<\/td>\n1<\/td>\n172000<\/td>\n0<\/td>\n<\/tr>\n
19<\/th>\n51<\/td>\n0<\/td>\n43000<\/td>\n1<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n","protected":false},"author":883,"menu_order":11,"template":"","meta":{"pb_show_title":"on","pb_short_title":"","pb_subtitle":"","pb_authors":[],"pb_section_license":""},"chapter-type":[],"contributor":[],"license":[],"class_list":["post-423","chapter","type-chapter","status-publish","hentry"],"part":64,"_links":{"self":[{"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/pressbooks\/v2\/chapters\/423","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/pressbooks\/v2\/chapters"}],"about":[{"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/wp\/v2\/types\/chapter"}],"author":[{"embeddable":true,"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/wp\/v2\/users\/883"}],"version-history":[{"count":3,"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/pressbooks\/v2\/chapters\/423\/revisions"}],"predecessor-version":[{"id":426,"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/pressbooks\/v2\/chapters\/423\/revisions\/426"}],"part":[{"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/pressbooks\/v2\/parts\/64"}],"metadata":[{"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/pressbooks\/v2\/chapters\/423\/metadata\/"}],"wp:attachment":[{"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/wp\/v2\/media?parent=423"}],"wp:term":[{"taxonomy":"chapter-type","embeddable":true,"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/pressbooks\/v2\/chapter-type?post=423"},{"taxonomy":"contributor","embeddable":true,"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/wp\/v2\/contributor?post=423"},{"taxonomy":"license","embeddable":true,"href":"https:\/\/pressbooks.bccampus.ca\/businessanalytics\/wp-json\/wp\/v2\/license?post=423"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}