Skip to main content

Table 2 Results of automatic labeling by FCNN, LR and SVM

From: Classifying the tracing difficulty of 3D neuron image blocks based on deep learning

Dataset

FCNN

LR

SVM

Accuracy

F1

Accuracy

F1

Accuracy

F1

Training

\(\underline{96.92\pm 0.19}\)

\(\underline{96.86\pm 0.20}\)

96.05±0.26

95.98±0.27

96.50±0.22

96.44±0.23

Test

\(\underline{96.49\pm 0.59}\)

\(\underline{96.40\pm 0.61}\)

96.07±0.53

95.98±0.54

96.30±0.57

96.22±0.59

  1. Numbers with underline are the best results among all models