SilasPenda/Nigerian-Food-Classification

This is an image classification project to classify Nigerian food to be either 'Yam', 'Egusi', 'Oha Soup', 'Afang Soup', 'Bread', 'Rice and Stew', 'Moi Moi', or 'Akara'. I created a streamlit web app to test the model. After the prediction, there is a link to a webpage where you can see recipes to make the predicted food. # Nigerian-Food-Classification This is an image classification project to classify Nigerian food to be either 'Yam', 'Egusi', 'Oha Soup', 'Afang Soup', 'Bread', 'Rice and Stew', 'Moi Moi', or 'Akara'. I created a streamlit web app to test the model. After the prediction, there is a link to a webpage where you can see recipes to make the predicted food. # Guide * Import necessary libraries * Read image directory and save images path in a list * Make a copy of the image paths list, shuffle and split into train, validation and test data * Generate new images using ImageDataGenerator * Load pretrained model and freeze the layers * Create layers * Train model * Test model and save as '.h5' model * Create web app using streamlit