python - How is "accuracy" calculated using Libsvm - SVM_Predict.exe -

-

i using libsvm first time. able train data(for images) , model ready "trainingdata.svm.model"

now, when run classification against unknown test data giving me 2 files: 1. trainingdata.svm.prediction (this file contains 1's , 0's) against each of test data. 2. giving me accuracy = 8 %

the question: 1. how interpret 1s , 0s in "trainingdata.svm.prediction". note: classifying genders 1 male , 0 female.

  1. how accuracy calculated? how can program calculate accuracy since test data unknown entity , not know labels yet.

thanks

  1. the file "trainingdata.svm.prediction" predicting labels 1 , 0 set (1 means sample predicted male, 0 female).

  2. it assumes labels belong class index 0, believe.


Comments

Post a Comment

Popular posts from this blog

javascript - Is getTimezoneOffset() stable during daylight saving transition? -

-
i have date time below conversion logic in javascript converting utc time passed timezone's local time. wondering logic work fine during daylight transition? if not remedy that? can't use third party library. have use pure javascript or angular js. function mytime() { var d1= document.getelementbyid("txtdate").value; var zoffset = document.getelementbyid("txtoffset").value; console.log("date1",d1); var d2 = new date(d1.replace(/ /g,'t')); var d3= d2.gettime()+(d2.gettimezoneoffset()*60000); console.log("date2",d2); console.log("date3",d3); var d4 = new date(d3 + (3600000 * zoffset)); console.log("date3",d3); console.log("date 4",d4); var d5 = d4.tolocaletimestring(); console.log("date 5",d5); ...
Read more

Why does PHP consider 0 to be equal to a string? -

-
i have following piece of code: $item['price'] = 0; /*code item information goes in here*/ if($item['price'] == 'e') { $item['price'] = -1; } it intended initialize item price 0 , information it. if price informed 'e' means exchange instead of sell, indicated negative value because stored in database requires numeric value. there possibility leave price 0, either because item bonus or because price set in later moment. but, when price not set, leaves initial value of 0, if loop indicated above evaluates true , price set -1. is, considers 0 equal 'e'. how can explained? edit: when price provided 0 (after initialization), behavior erratic: if evaluates true, evaluates false. you doing == sorts out types you. 0 int, in case going cast 'e' int. not parseable one, , become 0 . string '0e' become 0 , , match! use ===
Read more

formula - R: how to pass in a reference to the variable in glm or lm? -

-
so let's have named vector: sorted = c(1,2,3) names(sorted) = c("a","b","c") and it'll following: > sorted b c 1 2 3 so vector named a,b,c, , has value 1,2,3 respectively. and have sample data: data.ex = as.data.frame(matrix(rep(c(1,2,3,4),3), nrow = 3, ncol = 3)) colnames(data.ex) = c("a","b","c") so data frame has 3 columns named a,b,c well. i want predict c using value in glm(): fit.ex = glm(formula = c ~ names(sorted)[2], data = data.ex, family = binomial(link = "logit")) but then, i'll keep getting following error message: error in model.frame.default(formula = c ~ names(sorted)[2], data = data.ex,: variable lengths differ (found 'names(sorted)[2]') i read article here , found as.name() function, still not working: http://www.ats.ucla.edu/stat/r/pages/looping_strings.htm and cannot find else thats similar problem. please, if the...
Read more