print

README.md

@@ -42,6 +42,14 @@ models <- caretEnsemble::caretList(
   data = dat,
   methodList = c("rf", "glmnet")
 )
+print(summary(models))
+#> The following models were ensembled: rf, glmnet  
+#> 
+#> Model accuracy:
+#>    model_name metric    value        sd
+#>        <char> <char>    <num>     <num>
+#> 1:         rf   RMSE 1007.595 192.05082
+#> 2:     glmnet   RMSE 1217.374  69.54145
 ```
 
 Then, use caretEnsemble to make a greedy ensemble of these models
@@ -59,18 +67,18 @@ print(greedy_stack)
 #> Summary of sample sizes: 400, 400, 400, 400, 400 
 #> Resampling results:
 #> 
-#>   RMSE      Rsquared   MAE    
-#>   1034.614  0.9347528  517.897
+#>   RMSE      Rsquared  MAE     
+#>   951.4063  0.947765  545.1398
 #> 
 #> Tuning parameter 'max_iter' was held constant at a value of 100
 #> 
 #> Final model:
 #> Greedy MSE
-#> RMSE:  1047.523 
+#> RMSE:  959.2989 
 #> Weights:
 #>        [,1]
-#> rf     0.78
-#> glmnet 0.22
+#> rf     0.68
+#> glmnet 0.32
 ggplot2::autoplot(greedy_stack, training_data = dat, xvars = c("carat", "table"))
 ```
 
@@ -93,7 +101,7 @@ print(rf_stack)
 #> Resampling results:
 #> 
 #>   RMSE      Rsquared   MAE     
-#>   1107.429  0.9264549  532.1966
+#>   1030.494  0.9376481  541.5582
 #> 
 #> Tuning parameter 'mtry' was held constant at a value of 2
 #> 
@@ -105,8 +113,8 @@ print(rf_stack)
 #>                      Number of trees: 500
 #> No. of variables tried at each split: 2
 #> 
-#>           Mean of squared residuals: 1293471
-#>                     % Var explained: 92.1
+#>           Mean of squared residuals: 985743.6
+#>                     % Var explained: 93.99
 ggplot2::autoplot(rf_stack, training_data = dat, xvars = c("carat", "table"))
 ```
 

README.rmd

@@ -44,6 +44,7 @@ models <- caretEnsemble::caretList(
   data = dat,
   methodList = c("rf", "glmnet")
 )
+print(summary(models))
 ```
 
 Then, use caretEnsemble to make a greedy ensemble of these models

vignettes/Version-4.0-New-Features.Rmd

@@ -30,21 +30,21 @@ model_list <- caretEnsemble::caretList(
   y = iris[, 5L],
   methodList = c("rpart", "rf")
 )
-summary(model_list)
+print(summary(model_list))
 ```
 
 # Greedy Optimizer in caretEnsemble
 The new version uses a greedy optimizer by default, ensuring the ensemble is never worse than the worst single model:
 ```{r}
 ens <- caretEnsemble::caretEnsemble(model_list)
-summary(ens)
+print(summary(ens))
 ```
 
 # Enhanced S3 Methods
 caretStack (and by extension, caretEnsemble) now supports various S3 methods:
 ```{r}
 print(ens)
-summary(ens)
+print(summary(ens))
 plot(ens)
 ggplot2::autoplot(ens)
 ```

vignettes/caretEnsemble-intro.Rmd

@@ -42,7 +42,7 @@ model_list <- caretEnsemble::caretList(
   data = training,
   methodList = c("glmnet", "rpart")
 )
-summary(model_list)
+print(summary(model_list))
 ```
 (As with `train`, the formula interface is convenient but introduces move overhead. For large datasets the explicitly passing `x` and `y` is preferred).
 We can use the `predict` function to extract predictions from this object for new data:
@@ -63,7 +63,7 @@ model_list_big <- caretEnsemble::caretList(
     nn = caretEnsemble::caretModelSpec(method = "nnet", tuneLength = 2L, trace = FALSE)
   )
 )
-summary(model_list_big)
+print(summary(model_list_big))
 ```
 
 Finally, you should note that `caretList` does not support custom caret models. Fitting those models are beyond the scope of this vignette, but if you do so, you can manually add them to the model list (e.g. `model_list_big[["my_custom_model"]] <- my_custom_model`). Just be sure to use the same re-sampling indexes in `trControl` as you use in the `caretList` models!
@@ -84,7 +84,7 @@ caret::modelCor(caret::resamples(model_list))
 These 2 models make a good candidate for an ensemble: their predictions are fairly uncorrelated, but their overall accuracy is similar. We do a simple, linear greedy optimization on AUC using caretEnsemble:
 ```{r}
 greedy_ensemble <- caretEnsemble::caretEnsemble(model_list)
-summary(greedy_ensemble)
+print(summary(greedy_ensemble))
 ```
 
 ```{r}