Multi-Class Image Classification Model for Food 101 Identification Using TensorFlow Take 4

Template Credit: Adapted from a template made available by Dr. Jason Brownlee of Machine Learning Mastery.

SUMMARY: This project aims to construct a predictive model using a TensorFlow convolutional neural network (CNN) and document the end-to-end steps using a template. The Belgium Traffic Sign dataset is a multi-class classification situation where we attempt to predict one of several (more than two) possible outcomes.

INTRODUCTION: This dataset includes 101 food categories, totaling 101,000 images. The research team provided 250 manually reviewed test images and 750 training images for each food type. All images were rescaled to have a maximum side length of 512 pixels.

ANALYSIS: The VGG19 model’s performance achieved an accuracy score of 76.07% after ten epochs using the training dataset. When we applied the model to the validation dataset, the model achieved an accuracy score of 67.78%.

CONCLUSION: In this iteration, the TensorFlow VGG19 CNN model appeared suitable for modeling this dataset.

Dataset ML Model: Multi-Class classification with numerical features

Dataset Used: Food-101 – Mining Discriminative Components with Random Forests

Dataset Reference:

One source of potential performance benchmarks:

The HTML formatted report can be found here on GitHub.