Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

replace nround with nrounds to match actual parameter (#3592)

Browse Source
This commit is contained in:
Jakob Richter
2018-08-15 20:13:53 +02:00
committed by Philip Hyunsu Cho
parent 73bd590a1d
commit 725f4c36f2
17 changed files with 51 additions and 51 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
demo/kaggle-otto/otto_train_pred.R
View File

@@ -23,13 +23,13 @@ param <- list("objective" = "multi:softprob",
"nthread" = 8)
# Run Cross Validation
cv.nround = 50
cv.nrounds = 50
bst.cv = xgb.cv(param=param, data = x[trind,], label = y,
nfold = 3, nrounds=cv.nround)
nfold = 3, nrounds=cv.nrounds)
# Train the model
nround = 50
bst = xgboost(param=param, data = x[trind,], label = y, nrounds=nround)
nrounds = 50
bst = xgboost(param=param, data = x[trind,], label = y, nrounds=nrounds)
# Make prediction
pred = predict(bst,x[teind,])

10
demo/kaggle-otto/understandingXGBoostModel.Rmd
View File

@@ -121,19 +121,19 @@ param <- list("objective" = "multi:softprob",
"eval_metric" = "mlogloss",
"num_class" = numberOfClasses)
cv.nround <- 5
cv.nrounds <- 5
cv.nfold <- 3
bst.cv = xgb.cv(param=param, data = trainMatrix, label = y,
nfold = cv.nfold, nrounds = cv.nround)
nfold = cv.nfold, nrounds = cv.nrounds)
```
> As we can see the error rate is low on the test dataset (for a 5mn trained model).
Finally, we are ready to train the real model!!!
```{r modelTraining}
nround = 50
bst = xgboost(param=param, data = trainMatrix, label = y, nrounds=nround)
nrounds = 50
bst = xgboost(param=param, data = trainMatrix, label = y, nrounds=nrounds)
```
Model understanding
@@ -142,7 +142,7 @@ Model understanding
Feature importance
------------------
So far, we have built a model made of **`r nround`** trees.
So far, we have built a model made of **`r nrounds`** trees.
To build a tree, the dataset is divided recursively several times. At the end of the process, you get groups of observations (here, these observations are properties regarding **Otto** products).