I have trained a brf model for my data. when predict on the test Data, the result is wired as below. I split the data by createDataPartition() function. The performance for traindata also attatched.
ModelData.brf.train <- imbalanced(f, as.data.frame(traindata), method = "brf",
perf.type = "gmean", importance = TRUE)
print(ModelData.brf.test)
Sample size of test (predict) data: 18486
Number of grow trees: 3000
Average no. of grow terminal nodes: 2170.7517
Total no. of grow variables: 16
Resampling used to grow trees: swr
Resample size used to grow trees: 10566
Analysis: RF-C
Family: class
Imbalanced ratio: NaN
Brier score: 0.00144964
Normalized Brier score: 0.00579856
AUC: NaN
PR-AUC: NA
G-mean: NaN
Requested performance error: 0
Confusion matrix:
predicted
observed 0 1 class.error
0 0 0 NaN
1 0 0 NaN
Misclassification error: NaN
print(ModelData.brf.train)
Sample size: 69131
Frequency of class labels: 63848, 5283
Number of trees: 3000
Forest terminal node size: 1
Average no. of terminal nodes: 2170.7517
No. of variables tried at each split: 4
Total no. of variables: 16
Resampling used to grow trees: swr
Resample size used to grow trees: 10566
Analysis: RF-C
Family: class
Splitting rule: gini
Imbalanced ratio: 12.0856
(OOB) Brier score: 0.14548369
(OOB) Normalized Brier score: 0.58193477
(OOB) AUC: 0.81455718
(OOB) PR-AUC: 0.41409956
(OOB) G-mean: 0.73654094
(OOB) Requested performance error: 0.26345906
I have trained a brf model for my data. when predict on the test Data, the result is wired as below. I split the data by createDataPartition() function. The performance for traindata also attatched. ModelData.brf.train <- imbalanced(f, as.data.frame(traindata), method = "brf", perf.type = "gmean", importance = TRUE)
Confusion matrix:
observed 0 1 class.error 0 0 0 NaN 1 0 0 NaN
Confusion matrix:
observed 0 1 class.error 0 50647 13201 0.2068 1 1670 3613 0.3161