Book Image

Building Machine Learning Systems with Python - Third Edition

By : Luis Pedro Coelho, Willi Richert, Matthieu Brucher
Book Image

Building Machine Learning Systems with Python - Third Edition

By: Luis Pedro Coelho, Willi Richert, Matthieu Brucher

Overview of this book

Machine learning enables systems to make predictions based on historical data. Python is one of the most popular languages used to develop machine learning applications, thanks to its extensive library support. This updated third edition of Building Machine Learning Systems with Python helps you get up to speed with the latest trends in artificial intelligence (AI). With this guide’s hands-on approach, you’ll learn to build state-of-the-art machine learning models from scratch. Complete with ready-to-implement code and real-world examples, the book starts by introducing the Python ecosystem for machine learning. You’ll then learn best practices for preparing data for analysis and later gain insights into implementing supervised and unsupervised machine learning techniques such as classification, regression and clustering. As you progress, you’ll understand how to use Python’s scikit-learn and TensorFlow libraries to build production-ready and end-to-end machine learning system models, and then fine-tune them for high performance. By the end of this book, you’ll have the skills you need to confidently train and deploy enterprise-grade machine learning models in Python.
Table of Contents (17 chapters)
Free Chapter
1
Getting Started with Python Machine Learning

Deciding how to improve the performance

To improve on this, we basically have the following options:

  • Add more data: Maybe there is just not enough data for the learning algorithm; adding more training data should help.
  • Play with the model complexity: Maybe the model is not complex enough? Or maybe it is already too complex? In this case, we could decrease k so that it would take fewer nearest-neighbors into account and thus be better at predicting non-smooth data. Or we could increase it to achieve the opposite.
  • Modify the feature space: Maybe we do not have the right set of features? We could be missing some important aspect of the posts. Or should we remove some of our current features in case some features are aliasing others?
  • Change the model: Maybe kNN isn't a good fit for our use case; maybe it will never be capable of achieving good prediction performance, no matter...