eel4759_classification

Comparison of different ImageNet-based CNN models for classifying diabetic retinopathy images
Log | Files | Refs | README

classification_report.txt (593B)

      1 Experiment: resnet50_clahe=True
      2 
      3                       precision    recall  f1-score   support
      4 
      5          Healthy (0)       0.87      0.91      0.89      3872
      6        Mild NPDR (1)       0.18      0.16      0.17       366
      7    Moderate NPDR (2)       0.60      0.48      0.53       794
      8      Severe NPDR (3)       0.39      0.53      0.45       131
      9 Proliferative DR (4)       0.66      0.58      0.61       106
     10 
     11             accuracy                           0.78      5269
     12            macro avg       0.54      0.53      0.53      5269
     13         weighted avg       0.76      0.78      0.77      5269