Book Image

Python Deep Learning Projects

By : Matthew Lamons, Rahul Kumar, Abhishek Nagaraja
Book Image

Python Deep Learning Projects

By: Matthew Lamons, Rahul Kumar, Abhishek Nagaraja

Overview of this book

Deep learning has been gradually revolutionizing every field of artificial intelligence, making application development easier. Python Deep Learning Projects imparts all the knowledge needed to implement complex deep learning projects in the field of computational linguistics and computer vision. Each of these projects is unique, helping you progressively master the subject. You’ll learn how to implement a text classifier system using a recurrent neural network (RNN) model and optimize it to understand the shortcomings you might experience while implementing a simple deep learning system. Similarly, you’ll discover how to develop various projects, including word vector representation, open domain question answering, and building chatbots using seq-to-seq models and language modeling. In addition to this, you’ll cover advanced concepts, such as regularization, gradient clipping, gradient normalization, and bidirectional RNNs, through a series of engaging projects. By the end of this book, you will have gained knowledge to develop your own deep learning systems in a straightforward way and in an efficient way
Table of Contents (17 chapters)
8
Handwritten Digits Classification Using ConvNets

Conclusion

The first part of the project was to build an object detection classifier using YOLO architecture in Keras.

The second part of the project was to build a binary image segmentation model on COCO images that contain just a person, aside from the background. The goal was to build a good enough model to segment out the person from the background in the image.

The model we build by training on 1500 images, each of shape 360*480*3, has an accuracy of 79% on train data, and 78% on validation and test data. The model is successfully able to segment the person in the image, but the borders of the segmentations are slightly off from where they should be. This is due to using a small training set. Considering the number of images used for training, the model did a good job of segmenting.

There are more images available in this dataset that can be used for training, and it might...