You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
while building a model, as gbm skips scoring some trees (default behavior), it would be more useful to plot points instead of lines , bec gbm always reports the last point. Hence the user can extrapolate the behavior.
An example -
Building model using h2o-gbm and plotting using R-
tr = h2o.importFile(h,"/Users/nidhimehta/h2o/smalldata/gbm_test/ecology_model.csv",key = "tr")
myX = setdiff(colnames(tr),c("Angaus","Site"))
myY = "Angaus"
tru.gbms = h2o.gbm(x = myX, y = myY,training_frame=tr,loss = "gaussian",ntrees =400, nbins=20,min_rows=1,max_depth=4,learn_rate=.01,validation_frame= tr)
aa = tru.gbms@model$mse_valid
plot(aa,col = "red",xlab = "number of trees",ylab = "mean squared error", type ="p")
The text was updated successfully, but these errors were encountered:
while building a model, as gbm skips scoring some trees (default behavior), it would be more useful to plot points instead of lines , bec gbm always reports the last point. Hence the user can extrapolate the behavior.
An example -
Building model using h2o-gbm and plotting using R-
tr = h2o.importFile(h,"/Users/nidhimehta/h2o/smalldata/gbm_test/ecology_model.csv",key = "tr")
myX = setdiff(colnames(tr),c("Angaus","Site"))
myY = "Angaus"
tru.gbms = h2o.gbm(x = myX, y = myY,training_frame=tr,loss = "gaussian",ntrees =400, nbins=20,min_rows=1,max_depth=4,learn_rate=.01,validation_frame= tr)
aa = tru.gbms@model$mse_valid
plot(aa,col = "red",xlab = "number of trees",ylab = "mean squared error", type ="p")
The text was updated successfully, but these errors were encountered: