Residual Neural Network for image classification

Very deep neural networks suffer from a problem called vanishing/exploding gradients. In the year 2015 Kaiming He, et al proposed a solution for this in their research paper Deep Residual Learning for Image Recognition. A Residual Network, also known as ResNet, is a type of deep learning network architecture that introduces the concept of residual learning.

In this project I will be implementing a ResNet using ResNet50 architecture which consists of 50 layers.

Data Sources

This project is based on SIGNS dataset. This SIGNS dataset is a collection of 6 signs representing numbers from 0 to 5.

Technologies and Tools

• Tensorflow
• Keras

Installation

I have used TensorFlow framework for this project.

pip install tensorflow

🏆 Lessons Learned

1. Intuitions about Identity Block
2. Intuitions about Convolutional Block
3. Intuitions about ResNet50

Demo

Try it on my profile