Some machine learning software requires categorical data to be converted to numerical data. This section describes how to do this with one-hot encoding.<\/p>\n\n\n\n
Load the following small data frame:<\/p>\n\n\n\n
shoulders.rda<\/a><\/p>\n\n\n\n To look at the data, just enter:<\/p>\n\n\n\n The structure of the data frame can be seen with:<\/p>\n\n\n\n The first variable is categorical (factor) and the other two are numerical. The aim is to convert the categorical variable to numerical variables with hone-hot encoding. The package vtreat1<\/a><\/sup> should be installed<\/a>\u00a0 for the encoding and dplyr2<\/a><\/sup> to make filtering easier. Load the packages:<\/p>\n\n\n\n Create a vector that contains the names of the variables to be included in the final data frame (here all three):<\/p>\n\n\n\n Create a treatment plan (called treatplan) with vtreat’s designTreatmentsZ() function that takes as first argument the data frame to be treated (shoulders) and as second argument the variables to be included in the final data frame (vars):<\/p>\n\n\n\n The treatplan object contains a lot of information. The variables called “varName”, “origName” and “code” in the scoreFrame needs to be extracted and saved (here called scoreFrame):<\/p>\n\n\n\n Examine the scoreFrame:<\/p>\n\n\n\n The first variable (shoulder_lev_x.hemiarthroplasty)<\/em> is the first factor (hemiarthroplasty) of the categorical variable (shoulder) and the second variable (shoulder_lev_x.tsr<\/em>) is the second factor (tsr) of the categorical variable (shoulder). The third (fu_clean<\/em>) and fourth (score_clean<\/em>) variables are the cleaned up fu and score numerical variables.<\/p>\n\n\n\n We only want the rows with codes “clean” or “lev” (here all, but this is not always the case). Save them as a new object called new_variables:<\/p>\n\n\n\n Create a new data frame that contains the treated data and examine it:<\/p>\n\n\n\n The structure can be examined with the str() function:<\/p>\n\n\n\n The “shoulder” categorical variable has been recoded into new variables (binary) with as many variables as there are categories.<\/p>\n","protected":false},"excerpt":{"rendered":" Some machine learning software requires categorical data to be converted to numerical data. This section describes how to do this with one-hot encoding. Load the following small data frame: shoulders.rda To look at the data, just enter: The structure of the data frame can be seen with: The first variable is categorical (factor) and the […]<\/p>\n","protected":false},"author":2,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"inline_featured_image":false,"footnotes":""},"class_list":["post-2180","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/pcool.dyndns.org\/index.php\/wp-json\/wp\/v2\/pages\/2180","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/pcool.dyndns.org\/index.php\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/pcool.dyndns.org\/index.php\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/pcool.dyndns.org\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/pcool.dyndns.org\/index.php\/wp-json\/wp\/v2\/comments?post=2180"}],"version-history":[{"count":1,"href":"https:\/\/pcool.dyndns.org\/index.php\/wp-json\/wp\/v2\/pages\/2180\/revisions"}],"predecessor-version":[{"id":4624,"href":"https:\/\/pcool.dyndns.org\/index.php\/wp-json\/wp\/v2\/pages\/2180\/revisions\/4624"}],"wp:attachment":[{"href":"https:\/\/pcool.dyndns.org\/index.php\/wp-json\/wp\/v2\/media?parent=2180"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}shoulders<\/mark>\n shoulder fu score\n1 hemiarthroplasty 12 90\n2 hemiarthroplasty 10 80\n3 tsr 9 95\n4 hemiarthroplasty 8 89\n5 tsr 10 70<\/mark><\/em><\/code><\/pre>\n\n\n\nstr(shoulders)<\/em><\/span>\n'data.frame': 5 obs. of 3 variables:<\/em><\/span>\n $ shoulder: Factor w\/ 2 levels \"hemiarthroplasty\",..: 1 1 2 1 2<\/em><\/span>\n $ fu : num 12 10 9 8 10<\/em><\/span>\n $ score : num 90 80 95 89 70<\/em><\/span><\/code><\/pre>\n\n\n\nlibrary(vtreat)\u00a0<\/em><\/span>\nlibrary(dplyr)<\/em><\/span><\/code><\/pre>\n\n\n\nvars <- c(\"shoulder\", \"fu\", \"score\")<\/em><\/span>\nvars<\/em><\/span>\n[1] \"shoulder\" \"fu\" \"score\" <\/em><\/span><\/code><\/pre>\n\n\n\ntreatplan <- designTreatmentsZ(shoulders, vars)<\/mark>\n[1] \"vtreat 1.6.5 inspecting inputs Mon Jun 30 17:32:20 2025\"\n[1] \"designing treatments Mon Jun 30 17:32:20 2025\"\n[1] \" have initial level statistics Mon Jun 30 17:32:20 2025\"\n[1] \" scoring treatments Mon Jun 30 17:32:20 2025\"\n[1] \"have treatment plan Mon Jun 30 17:32:20 2025\"<\/mark><\/em><\/code><\/pre>\n\n\n\nscoreFrame <- treatplan$scoreFrame[ ,c(\"varName\", \"origName\", \"code\")]<\/em><\/span><\/code><\/pre>\n\n\n\nscoreFrame<\/mark>\n varName origName code\n1 fu fu clean\n2 score score clean\n3 shoulder_lev_x_hemiarthroplasty shoulder lev\n4 shoulder_lev_x_tsr shoulder lev<\/mark><\/em><\/code><\/pre>\n\n\n\nnew_variables <- dplyr::filter(scoreFrame, code %in% c(\"clean\", \"lev\"))<\/em><\/span>\nnew_variables<\/em><\/span>\n varName origName code\n1 fu fu clean\n2 score score clean\n3 shoulder_lev_x_hemiarthroplasty shoulder lev\n4 shoulder_lev_x_tsr shoulder lev<\/mark><\/em><\/code><\/pre>\n\n\n\nshoulder_treat <- prepare(treatplan, shoulders, varRestriction = new_variables$varName)<\/em><\/span>\nshoulder_treat<\/em><\/span>\n fu score shoulder_lev_x_hemiarthroplasty shoulder_lev_x_tsr\n1 12 90 1 0\n2 10 80 1 0\n3 9 95 0 1\n4 8 89 1 0\n5 10 70 0 1<\/mark><\/em><\/code><\/pre>\n\n\n\nstr(shoulder_treat)<\/mark>\n'data.frame':\t5 obs. of 4 variables:\n $ fu : num 12 10 9 8 10\n $ score : num 90 80 95 89 70\n $ shoulder_lev_x_hemiarthroplasty: num 1 1 0 1 0\n $ shoulder_lev_x_tsr : num 0 0 1 0 1<\/mark><\/em><\/code><\/pre>\n\n\n\n