r - Accuracy from Adaboost in caret not re-producable -


i using caret package train adaboost algorithm.

this training output

> adaboostfit1 bagged adaboost   507 samples  49 predictor   4 classes: 'a', 'b', 'c', 'd'   no pre-processing resampling: cross-validated (10 fold, repeated 10 times)  summary of sample sizes: 455, 456, 458, 458, 457, 456, ...  resampling results across tuning parameters:   maxdepth  mfinal  accuracy   kappa      1         50     0.7252999  0.5148544  1        100     0.7249040  0.5143589  1        150     0.7245041  0.5138308  1        200     0.7248961  0.5144177  3         50     0.8449199  0.7246524  3        100     0.8457983  0.7262998  3        150     0.8455672  0.7258904  3        200     0.8459633  0.7266197  5         50     0.8927122  0.8134603  5        100     0.8939299  0.8157804  5        150     0.8941498  0.8162610  5        200     0.8949189  0.8177514  7         50     0.8967421  0.8208284  7        100     0.8965234  0.8205643  7        150     0.8953196  0.8184594  7        200     0.8953154  0.8183581  9         50     0.8966646  0.8210096  9        100     0.8974849  0.8225458  9        150     0.8967082  0.8211391  9        200     0.8970887  0.8217614 11         50     0.8959179  0.8196260 11        100     0.8952882  0.8183883 11        150     0.8962304  0.8200345 11        200     0.8962304  0.8199575  kappa used select optimal model using  largest value. final values used model mfinal = 100 , maxdepth = 9.

the best model therefore 1 kappa of 0.8225458 , accuracy of 0.8974849. extract predictions using:

adaboostpre2<-predict(adaboostfit1)

and confusion matrix training data , perfect fit, when expecting 1 accuracy of 0.8975!

confusion matrix , statistics            reference prediction     b     c   d          265   0     0   0      b       0 186     0   0      c       0   0    11   0      d       0   0     0  45  overall statistics             accuracy : 1                        95% ci : (0.9928, 1) no information rate : 0.5227      p-value [acc > nir] : < 2.2e-16                  kappa : 1           mcnemar's test p-value : na           statistics class:                         class:   class: b     class: c   class: d sensitivity              1.0000     1.0000       1.0000    1.00000 specificity              1.0000     1.0000       1.0000    1.00000 pos pred value           1.0000     1.0000       1.0000    1.00000 neg pred value           1.0000     1.0000       1.0000    1.00000 prevalence               0.5227     0.3669       0.0217    0.08876 detection rate           0.5227     0.3669       0.0217    0.08876 detection prevalence     0.5227     0.3669       0.0217    0.08876 balanced accuracy        1.0000     1.0000       1.0000    1.00000

does have explanation why might case. related voting amonst various learners? trying other algorithms , c5.0 similary gives perfect fit. ideally access predictions gave accuracy of 0.8975. thanks.


Comments

Post a Comment

Popular posts from this blog

inversion of control - Autofac named registration constructor injection -

does autofac support specifying registration name in components' constructors? example: ninject's namedattribute . you need use autofac.extras.attributed package on top achieve this so let's have 1 interface , 2 classes: public interface ihello { string sayhello(); } public class englishhello : ihello { public string sayhello() { return "hello"; } } public class frenchhello : ihello { public string sayhello() { return "bonjour"; } } then have consumer class, in want select instance injected: public class helloconsumer { private readonly ihello helloservice; public helloconsumer([withkey("en")] ihello helloservice) { if (helloservice == null) { throw new argumentnullexception("helloservice"); } this.helloservice = helloservice; } public string sayhello() { return this.helloservice.sayhello()...
Read more

verilog - Systemverilog dynamic casting issues -

i've code snippet following in testbench function void write_to_port( my_data_type_base data ); my_data_type_extended data_ext; if(!$cast(data_ext, data)); `uvm_error(get_type_name(), "failed cast"); `uvm_info(get_name(), $psprintf("data_ext :\n%s", data_ext.sprint()), uvm_medium) // write data_ext out port.... endfunction when run it, i'm getting uvm_error "failed cast." i'm not quire sure why $cast not returning 1. can see, i'm printing out extended class data item after casting uvm_info. can see it's being cast properly. if don't use $cast if condition, don't runtime error. isn't coding practice use if dynamic cast check if $cast returning 1 ? what might reason behind cast not returning 1 in above case? i think semicolon on line 'if' not belong? i think consumes if statement, , uvm_error executes regardless of how if evaluates: if(!$cast(data_ext, data)); <- no semico...
Read more

ios - Change Storyboard View using Seague -

i have following code checks connection server, , depending on want change views if no connection found. nsstring *connect = [nsstring stringwithcontentsofurl:[nsurl urlwithstring:@"http://myserver.com"] encoding:nsutf8stringencoding error:nil]; if (connect == null) { [self performseguewithidentifier:@"network_failure" sender:self]; } else { // other code here } my problem not change views. have added storyboard segue between 2 views identifier of network_failure . not have navigation controller or that, problem? thanks help! no not problem. first of all, in objectivec exists nil object oriented null . must use that. so: if(connect == nil) or better: if(!connect) so due fact nil not same of null , probably code does't enter in if scope . you have been able discover breakpoint. anyway: be sure set trigger segue (line) 2 view controller , , name network_failure (that not best name. should instead view1toview2 ) as, m...
Read more