Classifying Defective Tires and Good Tires by using CNN and TL(MobileNetV2)
| Metrics | TL | CNN |
|---|---|---|
| Precision | 0.90 | 0.59 |
| Recall | 0.88 | 0.58 |
| F1-Score | 0.88 | 0.58 |
| Accuracy | 0.88 | 0.56 |
for using .h5
from keras.models import model_from_json
with open("name.json") as json_file:
model = model_from_json(json_file.read())
model.load_weights("name.h5")