Skip to main content

Table 7 Accuracy of the nine trained Resnet-101 individual models in classifying ALL in microscopic images when the best combination of hyperparameters was used in nine independent experimental runs

From: Classifying microscopic images as acute lymphoblastic leukemia by Resnet ensemble model and Taguchi method

Model Accuracy for the training set Accuracy for the preliminary test set
Resnet-101-8249(#1) 0.9881 0.8249
Resnet-101-8184(#2) 0.9856 0.8184
Resnet-101-8452(#3) 0.9872 0.8452
Resnet-101-8125(#4) 0.9893 0.8125
Resnet-101-8061(#5) 0.9841 0.8061
Resnet-101-8281(#6) 0.9848 0.8281
Resnet-101-8307(#7) 0.9811 0.8307
Resnet-101-8002(#8) 0.9877 0.8002
Resnet-101-8216(#9) 0.9859 0.8216
Average accuracy 0.9860 0.8209
SD 0.0025 0.0136
η value 37.0637 14.9359