Book Image

Modern Computer Vision with PyTorch

By : V Kishore Ayyadevara, Yeshwanth Reddy
Book Image

Modern Computer Vision with PyTorch

By: V Kishore Ayyadevara, Yeshwanth Reddy

Overview of this book

Deep learning is the driving force behind many recent advances in various computer vision (CV) applications. This book takes a hands-on approach to help you to solve over 50 CV problems using PyTorch1.x on real-world datasets. You’ll start by building a neural network (NN) from scratch using NumPy and PyTorch and discover best practices for tweaking its hyperparameters. You’ll then perform image classification using convolutional neural networks and transfer learning and understand how they work. As you progress, you’ll implement multiple use cases of 2D and 3D multi-object detection, segmentation, human-pose-estimation by learning about the R-CNN family, SSD, YOLO, U-Net architectures, and the Detectron2 platform. The book will also guide you in performing facial expression swapping, generating new faces, and manipulating facial expressions as you explore autoencoders and modern generative adversarial networks. You’ll learn how to combine CV with NLP techniques, such as LSTM and transformer, and RL techniques, such as Deep Q-learning, to implement OCR, image captioning, object detection, and a self-driving car agent. Finally, you'll move your NN model to production on the AWS Cloud. By the end of this book, you’ll be able to leverage modern NN architectures to solve over 50 real-world CV problems confidently.
Table of Contents (25 chapters)
Section 1 - Fundamentals of Deep Learning for Computer Vision
Section 2 - Object Classification and Detection
Section 3 - Image Manipulation
Section 4 - Combining Computer Vision with Other Techniques

Understanding the impact of learning rate annealing

So far, we have initialized a learning rate, and it has remained the same across all the epochs while training the model. However, initially, it would be intuitive for the weights to be updated quickly to a near-optimal scenario. From then on, they should be updated very slowly since the amount of loss that gets reduced initially is high and the amount of loss that gets reduced in the later epochs would be low.

This calls for having a high learning rate initially and gradually lowering it later on as the model achieves near-optimal accuracy. This requires us to understand when the learning rate must be reduced.

One potential way we can solve this problem is by continually monitoring the validation loss and if the validation loss does not decrease (let's say, over the previous x epochs), then we reduce the learning rate.

PyTorch provides us with tools we can use to perform learning rate reduction when the validation loss does not...