Skip to main content
Fig. 2 | BMC Genetics

Fig. 2

From: funbarRF: DNA barcode-based fungal species prediction using multiclass Random Forest supervised learning model

Fig. 2

(a) Line graphs showing the trend of OOB-error rates with respect to different number of classification trees (ntree) in RF. b The OOB-error rates for different model representations with default values of mtry at ntree=500. c Heat maps of the OOB-error rates at ntree=500 with different values of mtry for different model representations. d Heat map of the OOB-error rate for the dataset with 9 sequences per species for different mtry values and model representations. It can be seen that the OOB-error got stabilized after reaching 400 classification trees, whereas mtry=\( \sqrt{\mathrm{p}}\ \Big( \)9) was observed optimum due to less OOB-error rates as compared to the other values of mtry

Back to article page

BMC Genomic Data

ISSN: 2730-6844

Contact us