Table 2 Test accuracy of climacteric fruits using distinct models.
From: Zero-shot transfer learned generic AI models for prediction of optimally ripe climacteric fruits
Base model | Trained on | Tested on | Model name | Test accuracy (TA) |
|---|---|---|---|---|
Fine-tuned VGG-16 | Banana | Banana | Model-1 | 66.7 |
Fine-tuned VGG-16 | Papaya | Papaya | Model-2 | 70.7 |
Fine-tuned VGG-16 | Mango | Mango | Model-3 | 72.3 |
Fine-tuned VGG-16 | Lemon | Lemon | Model-4 | 79.6 |
Fine-tuned VGG-16 | Banana and papaya | Banana and papaya | Model-5 | 78.1 |
Fine-tuned VGG-16 | Mango and banana | Mango and banana | Model-6 | 80.2 |
Fine-tuned VGG-16 | Mango and papaya | Mango and papaya | Model-7 | 83.3 |
Fine-tuned VGG-16 further fine-tune on ‘Fruits 360’ dataset | Banana | Banana | Model-8 | 83.3 |
Fine-tuned VGG-16 further fine-tune on ‘Fruits 360’ dataset | Papaya | Papaya | Model-9 | 87.6 |
Fine-tuned VGG-16 further fine-tune on ‘Fruits 360’ dataset | Mango | Mango | Model-10 | 84.6 |
Fine-tuned VGG-16 further fine-tune on ‘Fruits 360’ dataset | Lemon | Lemon | Model-11 | 86.4 |
Fine-tuned VGG-16 further fine-tune on ‘Fruits 360’ dataset | Banana and papaya | Banana and papaya | Model-12 | 87.5 |
Fine-tuned VGG-16 further fine-tune on ‘Fruits 360’ dataset | Mango and banana | Mango and banana | Model-13 | 88.2 |
Fine-tuned VGG-16 further fine-tune on ‘Fruits 360’ dataset | Mango and papaya | Mango and papaya | Model-14 | 89.9 |
