View source: R/ranger_reg_plot.R
plot_test_perf_VS_rand | R Documentation |
This outputs a histogram and a p-value showing if the performance of a real regression model significantly better than null models.
plot_test_perf_VS_rand( rf_model, newx, newy, prefix = "test", target_field = "", metric = "MAE", permutation = 1000, n_features = NA, outdir = NULL )
rf_model |
A trained rf model object, which should be generated from /coderf.cross.validation or /coderf.out.of.bag. |
newx |
A data.matrix or data.frame with the new data for rf model testing. |
newy |
The data label of new data. |
prefix |
The prefix for the dataset in the training or testing. |
target_field |
A string indicating the target field of the metadata for machine-learning analysis. |
metric |
The regression performance metric applied, including MAE, RMSE, MSE, R_squared, Adj_R_squared, or Separman_rho. |
permutation |
The permutation times for a random guess of regression performance. |
n_features |
The number of features in the training data. |
outdir |
The output directory. |
Shi Huang
set.seed(123) x <- data.frame(rbind(t(rmultinom(7, 75, c(.201,.5,.02,.18,.099))), t(rmultinom(8, 75, c(.201,.4,.12,.18,.099))), t(rmultinom(15, 75, c(.011,.3,.22,.18,.289))), t(rmultinom(15, 75, c(.091,.2,.32,.18,.209))), t(rmultinom(15, 75, c(.001,.1,.42,.18,.299))))) y<- 1:60 newx <- data.frame(rbind(t(rmultinom(7, 75, c(.201,.5,.02,.18,.099))), t(rmultinom(8, 75, c(.201,.4,.12,.18,.099))), t(rmultinom(15, 75, c(.001,.1,.42,.18,.299))))) newy<- 4:33 rf_model<-rf.out.of.bag(x, y) p<-plot_perf_VS_rand(x=x, y=y, predicted_y=rf_model$predicted, prefix="train", nfolds=5, permutation=100, metric="MAE", target_field="age", n_features=5) p p_test<-plot_test_perf_VS_rand(rf_model, newx, newy, permutation=100, metric="MAE", target_field="age", n_features=5) p_test
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.