Transfer-Learning-App-TensorfowLite-and-Flutter

---

1. Built and Trained a Convolutional Neural Network using Transfer Learning in Tensorflow and Keras inside of Jupyter Notebook.

The script is stored inside the pythonscripts.ipynb file.

2. Converted and Exported it as a .tflite asset into a blank Flutter Project

All app files are in the transfer_learning_fruit_veggies folder.

3. Finally, wrote a fully functional Flutter mobile app that uses plugins such as tflite and image_picker to use on-device machine learning and now is able to detect 30+ different types of fruits and vegetables, from either a photo taken in real time or an image selected from a camera roll with 97.76% training accuracy and 78.56% validation accuracy

Test this project out yourself, by cloning the repository into Visual Studio Code