NutriPinoy: A Computer Vision-Based Classifier for Filipino Food and Nutritional Facts

Prerequisite

• Python 3.9.18
• Anaconda (Make sure you are in an environment with the correct python version)

Install Dependencies

• Run conda create -n python=3.9.18 yourenv pip to create an environment with pip installed.
• Run conda activate yourenv to activate your virtual environment.
• (While at the root of the project directory) Run pip install -r requirements.txt to install the necessary dependencies.

Model Training Guide

• In the model_training directory, you can find the image_data folder. This is where the dataset is stored. You can view it and add your own data. Just make sure to follow the same folder structure.

• You can run the train_classifier.ipynb to train your own data. You can also change some parameters, such as the epoch, image dimension, etc., in the Jupyter notebook.

• Important: After running the Jupyter notebook, two files will be generated: resnet_model.h5 and class_names.json. If you train your own data, you need to relocate resnet_model.h5 to the root of the application folder and transfer class_names.json to /applications/configurations

Running the application

Reminders: Before running the application, ensure that you have the generated resnet_model.h5 from the model training on the root directory of the application folder. Moreover, make sure that class_names.json is available in the configurations folder.

Activate your anaconda environment first by inputting the following command into the terminal:

conda activate yourenv

Go to the application directory in the terminal through (assuming your terminal working directory is the root directory of the project folder):

cd application

To run the application, input the following command in the terminal:

python main.py

Useful Information

File Structure

• application: Contains the files for running the application
• model_training: Contains the files for training the ResNet50 image classification model.

Application Directories

• The utils directory contains the helper classes.
• The configurations folder contains the class which for the application configurations which you can modify.

Application Classes

• UseApi: Class for calling the API Ninjas which is used to obtain data about nutritional facts
• UseModel: Class to load the image and make predictions using the trained model.
• ModelConfig: Class for modifying applications configurations, such as the dimensions of the image.