<![CDATA[This study aims to classify various types of flowers using the deep learning method with the Convolutional Neural Networks MobileNetV3-Small architecture. The research stages include data collection and sharing, model training using the MobileNetV3-Small architecture, and testing with evaluation using the Confusion Matrix. Eight models with hyperparameter variations were tested to find the model with the highest accuracy. Model five achieved the highest validation accuracy of 99.25%. The evaluation showed that model five achieved the highest accuracy of 92%. These results indicate that increasing the amount of data and parameter settings can significantly improve model accuracy, indicating that the model can achieve near-perfect performance with more training data and optimal parameter settings. This study uses open-access data from Kaggle covering five types of flowers: daisy, dandelion, rose, sunflower, and tulip.]]>
Copyrights © 2024
