Machine Learning documentation

The process :

1. Loading the Dataset
2. Loading Pre-trained Model (MobileNetV2)
3. Tuning the Model
4. Building the Model
5. Training the Model
6. Saving the Model
7. Model Evaluation
8. Converting Model to TFLite

Using MobileNet V2

We using MobileNetV2 for transfer learning in image classification tasks in this application because has several advantages:

• MobileNetV2 is designed to be lightweight and efficient
• MobileNetV2 utilizes depthwise separable convolution
• MobileNetV2 has demonstrated good performance on mobile and edge device
• MobileNetV2 is pre-trained on large-scale datasets, often with millions of images
• TensorFlow Hub provides a convenient way to access pre-trained models, including MobileNetV2, for transfer learning