Tensorflow implementation of "Compounding the Performance Improvements of Assembled Techniques in a Convolutional Neural Network"

A system that takes food images as an input, recognizes the food automatically and gives the nutritional-facts as an output.

Building a model for predicting food images using the famous FOOD-101 dataset.

training food-101 (achieved SOTA top-1 validation acc ~=90%) using 1-cycle-policy:

Vision Transformer Based Food-101 Classification

This repo contains implementations of the challenges from fellowship.ai. For more, visit here.

Web application using Streamlit and Keras to predict food class.

Food classification on Food-101 dataset with optimization of training with TF Profiler, transfer learning, fine-tuning, intelligibility

Training a Deep neural network with torch- Application on food recognition

A deep learning project for food image classification using the Food-101 dataset, leveraging DenseNet201 architecture. Includes nutritional facts estimation with scope for improvement in accuracy and scalability.

classifying the food101 dataset using cnn

Using Keras models and datasets to build custom prediction models

Fine-tuning a MobileNet model on the Food-101 dataset. Involves experimenting with different techniques for optimal performance, and features a Flask web application for real-time inference.

Example application for training Microsofts's pretrained BEiT image transformer model on a new image classification task