Get SH*T Done with PyTorch: Solve Real-World Machine Learning Problems with Deep Neural Networks in Python

by Venelin Valkov

PyTorch is the best Deep Learning library there (currently) is, period! Doing ML with PyTorch feels like a superpower (of course, there are bad parts, too). Trust me, I have a book on TensorFlow and Keras! This opinion comes from my real-world experience, as a Machine Learning Engineer, and writer of numerous Machine Learning and Deep Learning tutorials. This book skips the bull and goes straight into solving real-world problems. There is some theory, only where you need to connect the dots. You'll go from the basics of using PyTorch to solving Computer Vision, Natural Language, and Time Series problems with complete source code and runnable Jupyter notebooks. The examples are compatible with the latest versions of PyTorch and Torchvision. Here’s what you’ll learn from this - Getting Started with PyTorch - Build Your First Neural Network with PyTorch - Transfer Learning for Image Classification using Torchvision - Time Series Forecasting with LSTMs for Daily Coronavirus Cases - Time Series Anomaly Detection using LSTM Autoencoders - Face Detection on Custom Dataset with Detectron2 - Create Dataset for Sentiment Analysis by Scraping Google Play App Reviews - Sentiment Analysis with BERT and Transformers by Hugging Face - Deploy BERT for Sentiment Analysis as REST API using FastAPI

196 pages, Kindle Edition

Published May 25, 2020

Reviews

[Enlik Lee · Rating 5 out of 5]

A lot of hands-on tutorial with link to GitHub and google collab that made this book feels like a well-documented tutorial.
Every chapter contains a specific topic about how PyTorch can be implemented in many different real-world cases.